Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
667 commits
Select commit Hold shift + click to select a range
e597ec3
feat(modnet): refine pi review input and quit behavior
EdwardIrby Mar 26, 2026
2651c4d
docs(agent): add evolutionary harness research program
EdwardIrby Mar 26, 2026
1292e8f
chore(modnet): move training prompts program
EdwardIrby Mar 27, 2026
14ebe8f
feat(research): add hypergraph encoding tooling
EdwardIrby Mar 28, 2026
146b94e
feat(mss): rename skill and audit agent-era semantics
EdwardIrby Mar 28, 2026
6ed7e46
feat(research): add MSS source comparison runner
EdwardIrby Mar 28, 2026
347036e
chore(research): prefer free Nemotron embeddings for MSS compare
EdwardIrby Mar 28, 2026
3fadfcd
feat(research): tighten program-runner split for fanout
EdwardIrby Mar 28, 2026
1c28f02
feat(research): tighten lane script boundaries
EdwardIrby Mar 28, 2026
d73aea2
fix(research): stabilize fanout execution checks
EdwardIrby Mar 28, 2026
9d9fe61
feat(research): harden default hypergraph promotion flow
EdwardIrby Mar 28, 2026
fa9dca9
refactor(default-hypergraph): simplify seed contract
EdwardIrby Mar 28, 2026
a210cc1
feat(research): add mss seed autoresearch lanes
EdwardIrby Mar 28, 2026
f08c884
chore(research): trim stale MSS helper commands
EdwardIrby Mar 28, 2026
493e96c
chore(research): align MSS commands with autoresearch
EdwardIrby Mar 28, 2026
6e6ab45
refactor(research): inline embedding transport into MSS compare
EdwardIrby Mar 28, 2026
9b91e23
feat(research): add lane-specific prompts for MSS runs
EdwardIrby Mar 28, 2026
07f0c99
feat(research): add workspace improvement lane evaluation
EdwardIrby Mar 28, 2026
a1f084d
docs: refresh readme and archive native model proof plan
EdwardIrby Mar 28, 2026
748b69a
fix(research): anchor MSS autoresearch writes to worktrees
EdwardIrby Mar 28, 2026
943f34e
feat(research): harden autoresearch evaluation flow
EdwardIrby Mar 28, 2026
e758fda
feat(mss-seed): commit generated seed artifacts
EdwardIrby Mar 28, 2026
fa85d3b
feat(mss-corpus): generate corpus artifacts from source material
EdwardIrby Mar 28, 2026
6ead672
chore(research): reset lane outputs and runner entrypoints
EdwardIrby Mar 28, 2026
5ff1c35
fix(research): tighten pi worktree sandbox policy
EdwardIrby Mar 28, 2026
f0489a4
feat(research): add behavioral memory lanes
EdwardIrby Mar 28, 2026
cebc2e6
refactor(modnet): fold implementation doc into skills
EdwardIrby Mar 28, 2026
333955d
refactor(research): split evolving skills into programs
EdwardIrby Mar 28, 2026
b9cec13
refactor(research): align improve and behavioral lanes
EdwardIrby Mar 28, 2026
fc4b9b5
chore(research): capture mss-seed attempt 05
EdwardIrby Mar 28, 2026
1e45a03
chore(research): capture behavioral-seed attempt 05
EdwardIrby Mar 28, 2026
3858953
chore(research): capture behavioral-corpus attempt 01
EdwardIrby Mar 29, 2026
859cefd
chore(research): capture mss-corpus attempt 18
EdwardIrby Mar 29, 2026
d905bd6
refactor(research): collapse legacy lanes into factories
EdwardIrby Mar 29, 2026
68af76f
chore: remove artifacts
EdwardIrby Mar 29, 2026
4b2155d
fix(research): harden behavioral factories evaluation
EdwardIrby Mar 29, 2026
82e7eb0
docs(research): rewrite behavioral factories program
EdwardIrby Mar 29, 2026
6f7dcae
docs(research): clarify behavioral factories architecture
EdwardIrby Mar 29, 2026
f534fc1
docs(research): align factories vocabulary and agent shape
EdwardIrby Mar 29, 2026
c797819
docs(research): sync agent and runtime refactor notes
EdwardIrby Mar 30, 2026
ec93aef
refactor(agent): move bootstrap and link ownership
EdwardIrby Mar 30, 2026
045add7
refactor(runtime): remove team orchestration surfaces
EdwardIrby Mar 30, 2026
66bb2f7
feat(agent): add minimal agent core surface
EdwardIrby Mar 30, 2026
173b496
refactor(inference): move openai compat model adapter
EdwardIrby Mar 30, 2026
b112260
test(inference): remove stale agent openai compat test
EdwardIrby Mar 30, 2026
e1c3bb4
feat(agent): add runtime factory update event
EdwardIrby Mar 30, 2026
9929b26
docs: clarify factory BP constraints and module boundaries
EdwardIrby Mar 30, 2026
4670fa8
feat(factories): add default inference factory
EdwardIrby Mar 30, 2026
7bbb3f9
feat(factories): add gate and execute factory
EdwardIrby Mar 30, 2026
d8488e9
feat(factories): add simulation and evaluation factory
EdwardIrby Mar 30, 2026
71c02fa
refactor(improve): drop distillation adapter path
EdwardIrby Mar 30, 2026
7b1b583
docs(research): capture create-node loop blocker
EdwardIrby Mar 30, 2026
44aaf7c
refactor(bootstrap): remove modnet transitional layer
EdwardIrby Mar 30, 2026
8c535a2
docs(research): capture bootstrap and a2a boundaries
EdwardIrby Mar 30, 2026
b5ab94b
refactor(cli): remove transitional acp entrypoint
EdwardIrby Mar 30, 2026
86b19ba
docs(training-prompts): clarify acp transport direction
EdwardIrby Mar 30, 2026
8582486
refactor(agent): remove legacy create-agent-loop surface
EdwardIrby Mar 30, 2026
edf5c09
refactor(factories): inline simulation evaluation helper
EdwardIrby Mar 30, 2026
ed4df18
refactor(agent): drop legacy executor helpers
EdwardIrby Mar 30, 2026
b43d64d
refactor(improve): drop distillation adapter path
EdwardIrby Mar 30, 2026
4728e61
refactor(agent): move simulation logic into factory
EdwardIrby Mar 30, 2026
32e2978
refactor(agent): move snapshot capture into runtime sql
EdwardIrby Mar 30, 2026
2f67efb
refactor(agent): remove legacy proactive surface
EdwardIrby Mar 30, 2026
e4f4fce
refactor(agent): remove dead context assembly layer
EdwardIrby Mar 30, 2026
8c2c32c
refactor(agent): absorb link runtime into agent
EdwardIrby Mar 30, 2026
2881d80
docs(research): sync factory refactor notes
EdwardIrby Mar 30, 2026
c12c07f
refactor(agent): move CRUD capability into core
EdwardIrby Mar 30, 2026
284f1f3
feat(agent): add core tool execution events
EdwardIrby Mar 30, 2026
713a627
refactor(factories): route gate execute through agent core tools
EdwardIrby Mar 30, 2026
132d752
refactor(agent): absorb truncation helper
EdwardIrby Mar 30, 2026
4470f76
refactor(tools): drop dead validation and model wrappers
EdwardIrby Mar 30, 2026
002523f
refactor(tools): remove stale thread validator
EdwardIrby Mar 30, 2026
60cb9f7
refactor(skill): move skill exports out of tools
EdwardIrby Mar 30, 2026
a5f029c
refactor(tools): remove research validator
EdwardIrby Mar 30, 2026
fa9c7e7
refactor(hypergraph): promote hypergraph module boundary
EdwardIrby Mar 30, 2026
b73587e
refactor(cli): move lsp and hypergraph boundaries
EdwardIrby Mar 30, 2026
6245704
refactor(tools): remove remaining tools directory
EdwardIrby Mar 30, 2026
4087a04
docs(research): sync factory notes with current boundaries
EdwardIrby Mar 30, 2026
097f38a
fix(exports): restore main cli re-export
EdwardIrby Mar 30, 2026
8ab8dea
refactor(cli): centralize current package command surfaces
EdwardIrby Mar 30, 2026
33a4e0d
refactor(cli): rename trial command to eval
EdwardIrby Mar 30, 2026
a86b2ef
refactor(improve): rename trial support boundary to eval
EdwardIrby Mar 30, 2026
4a6f36d
refactor(improve): remove training annotation from eval outputs
EdwardIrby Mar 30, 2026
196849e
docs(research): align factory notes with eval boundary
EdwardIrby Mar 30, 2026
c22d7d4
refactor(improve): align tooling with evolutionary lane
EdwardIrby Mar 30, 2026
77931db
refactor(improve): remove training lane
EdwardIrby Mar 30, 2026
09e367a
docs(research): sync improve boundary with evolutionary lane
EdwardIrby Mar 30, 2026
dd510c7
docs(notes): add next session checkpoint
EdwardIrby Mar 30, 2026
20dc23f
refactor(improve): promote meta-verification to top level
EdwardIrby Mar 31, 2026
9bb75e5
docs(research): reframe evolutionary unit around Plaited node
EdwardIrby Mar 31, 2026
20f6f28
docs(research): add tool replay direction to evolutionary lane
EdwardIrby Mar 31, 2026
5217db9
feat(training-prompts): bucket prompt catalog
EdwardIrby Apr 1, 2026
e27f935
feat(training-prompts): curate buckets and scope pi review
EdwardIrby Apr 1, 2026
c5522de
fix(behavioral): remove redundant restricted trigger cast
EdwardIrby Apr 1, 2026
f7ac262
refactor(agent): simplify core and align factory contract
EdwardIrby Apr 1, 2026
b903f26
refactor(agent): route signal registration through bp
EdwardIrby Apr 1, 2026
8373adc
refactor(agent): scope factory loading to workspace
EdwardIrby Apr 1, 2026
067d3a3
refactor(src): simplify agent runtime surfaces
EdwardIrby Apr 2, 2026
87394b0
docs(research): sync behavioral factory architecture
EdwardIrby Apr 2, 2026
7ba6da1
feat(training-prompts): audit bucket prompts with mss judge
EdwardIrby Apr 2, 2026
a4f0c4b
refactor: remove pi review surface and align behavioral docs
EdwardIrby Apr 2, 2026
273881e
refactor(agent): unify signal-backed tool results
EdwardIrby Apr 2, 2026
9da5dce
docs(research): prune deleted dev-research lanes
EdwardIrby Apr 2, 2026
df07620
refactor(agent): remove legacy link and util surfaces
EdwardIrby Apr 2, 2026
f8c3f38
docs(research): remove sub-agent architecture references
EdwardIrby Apr 2, 2026
faa0508
refactor(agent): tighten signal helper surface
EdwardIrby Apr 2, 2026
4b78dcf
refactor: remove default factory runtime surface
EdwardIrby Apr 2, 2026
2e04cd5
refactor(agent): align module factory loading
EdwardIrby Apr 2, 2026
626c75c
docs(skills): sync agent skill links
EdwardIrby Apr 2, 2026
fc8a7e2
refactor: reorganize skill loading surfaces
EdwardIrby Apr 3, 2026
e585717
refactor: move cli into utils
EdwardIrby Apr 3, 2026
d47b115
refactor(agent): tighten runtime guard schemas
EdwardIrby Apr 3, 2026
df3774e
docs(research): add skill factories lane
EdwardIrby Apr 3, 2026
1332099
chore(factories): add factory export surface
EdwardIrby Apr 3, 2026
da5f417
refactor(factories): consolidate default factory research and surfaces
EdwardIrby Apr 3, 2026
a27adf2
docs(research): add memory factories lane
EdwardIrby Apr 3, 2026
528704a
docs: add infrastructure deployment guide
EdwardIrby Apr 3, 2026
94e4fd5
docs: clean up src tsdoc comments
EdwardIrby Apr 3, 2026
6484fc1
feat: add bootstrap cli and deployment surfaces
EdwardIrby Apr 3, 2026
2cc9193
docs: remove stale training and modules doctrine
EdwardIrby Apr 3, 2026
94f32d2
refactor: remove stale research script surfaces
EdwardIrby Apr 3, 2026
fe5340f
docs: consolidate plaited ui skill
EdwardIrby Apr 3, 2026
b5e76b7
feat: add factory program runner
EdwardIrby Apr 3, 2026
0d6e82b
docs: consolidate improve workflow skills
EdwardIrby Apr 3, 2026
da6bd6f
docs: prune stale agent loop skills
EdwardIrby Apr 3, 2026
e3e7f4d
docs(research): add acp factory lane
EdwardIrby Apr 3, 2026
b7d0e8d
docs: consolidate modnet skill surfaces
EdwardIrby Apr 3, 2026
ad7200e
docs(research): add factory research lanes
EdwardIrby Apr 3, 2026
5614b58
docs(research): rename harness research lane
EdwardIrby Apr 3, 2026
a6a10ab
chore(skills): refresh workspace skill symlinks
EdwardIrby Apr 3, 2026
e8358f4
docs(research): add search factory lane
EdwardIrby Apr 3, 2026
5ee0074
docs(research): add verification factory lane
EdwardIrby Apr 3, 2026
87edb0f
docs: define node-home persistence layering
EdwardIrby Apr 4, 2026
bf786ff
docs(research): add workflow and payment factory lanes
EdwardIrby Apr 4, 2026
5c8e53d
docs: define node discovery and MSS extension policy
EdwardIrby Apr 4, 2026
d86cab8
docs(research): add factory research lanes
EdwardIrby Apr 4, 2026
944cf32
docs(research): add identity trust lane
EdwardIrby Apr 4, 2026
e89c76d
refactor(agent): drop legacy loop constants
EdwardIrby Apr 4, 2026
4531f7c
refactor(improve): rename distillation retention labels
EdwardIrby Apr 4, 2026
9819830
refactor(agent): rename core event surface
EdwardIrby Apr 4, 2026
0b1c039
refactor(agent): remove unused tool status constant
EdwardIrby Apr 4, 2026
190cd63
refactor(improve): reduce program runner surface
EdwardIrby Apr 4, 2026
d4b9c96
refactor: split improve into eval and program-runner
EdwardIrby Apr 5, 2026
427b0b9
refactor(agent): align Gemma 4 autoresearch lane
EdwardIrby Apr 5, 2026
19f62d4
refactor: narrow package exports and align module surfaces
EdwardIrby Apr 5, 2026
414f650
refactor(ui): localize protocol event constants
EdwardIrby Apr 5, 2026
0a0237c
feat(skills): add code documentation audit cli
EdwardIrby Apr 5, 2026
af9d934
feat(skills): expand typescript lsp audit operations
EdwardIrby Apr 5, 2026
e420cb9
feat(research): add autoresearch cli scaffold
EdwardIrby Apr 5, 2026
bbc113b
refactor(skills): align autoresearch workflow tooling
EdwardIrby Apr 5, 2026
8d5bd24
feat(skills): add audit and remote MCP cli
EdwardIrby Apr 5, 2026
5e2ba6e
chore(scripts): remove legacy script entrypoints
EdwardIrby Apr 5, 2026
e8ff50d
feat(autoresearch): add autonomous factory orchestration
EdwardIrby Apr 5, 2026
3292786
chore: gitignore update
EdwardIrby Apr 5, 2026
58fb1a5
fix(autoresearch): enforce program-runner lane scope
EdwardIrby Apr 5, 2026
779c800
chore: update:
EdwardIrby Apr 5, 2026
9276620
fix(autoresearch): use Pi SDK worker defaults
EdwardIrby Apr 5, 2026
7aa0d6d
fix(autoresearch): avoid inline lane prompt stalls
EdwardIrby Apr 5, 2026
cb478eb
fix(autoresearch): strengthen judge context
EdwardIrby Apr 5, 2026
ac4ffe6
feat(autoresearch): split planning from execution
EdwardIrby Apr 5, 2026
e6bf697
fix(autoresearch): invoke codex planner correctly
EdwardIrby Apr 5, 2026
439d8cc
fix(program-runner): honor explicit run dir
EdwardIrby Apr 5, 2026
935cd13
fix(autoresearch): tighten execution evidence
EdwardIrby Apr 5, 2026
6b3b18d
fix(autoresearch): tighten planning and tool diagnostics
EdwardIrby Apr 5, 2026
b14641e
chore(autoresearch): switch execution default to qwen free
EdwardIrby Apr 5, 2026
72e2d40
chore(autoresearch): set kimi as execution fallback
EdwardIrby Apr 5, 2026
b732544
fix(autoresearch): gate promotion through codex
EdwardIrby Apr 5, 2026
686e276
fix(autoresearch): timeout stalled planner runs
EdwardIrby Apr 5, 2026
d51b30a
chore(skills): sync agent skill symlinks
EdwardIrby Apr 5, 2026
39f3f22
fix(autoresearch): isolate planner from injected secrets
EdwardIrby Apr 5, 2026
5c243a0
fix(autoresearch): isolate promotion from injected secrets
EdwardIrby Apr 5, 2026
b559e90
fix(autoresearch): move planning and promotion fully to pi
EdwardIrby Apr 5, 2026
c44b9a5
fix(autoresearch): split primary and fallback execution timeouts
EdwardIrby Apr 5, 2026
d8dd5b0
fix(autoresearch): tighten lane guidance for two-program smoke
EdwardIrby Apr 5, 2026
c192c45
fix(autoresearch): parallelize program lanes
EdwardIrby Apr 5, 2026
fab21ce
chore: revert
EdwardIrby Apr 5, 2026
9629604
chore: revert
EdwardIrby Apr 5, 2026
51f6409
refactor(autoresearch): split lane control from agent adapters
EdwardIrby Apr 6, 2026
c893779
feat(factories): build base factory lanes
EdwardIrby Apr 6, 2026
2dcbd29
chore: neutralize behavioral core terminology
EdwardIrby Apr 6, 2026
b9570aa
refactor(ui): centralize protocol event maps
EdwardIrby Apr 6, 2026
2ccd3b4
chore(ui): commit staged ui updates
EdwardIrby Apr 6, 2026
7ae1ac4
revert(ui): restore factory wording
EdwardIrby Apr 6, 2026
082bd85
chore(behavioral): remove remaining factory wording
EdwardIrby Apr 6, 2026
7e93c71
Revert "chore: neutralize behavioral core terminology"
EdwardIrby Apr 6, 2026
0ccb23d
--wip-- [skip ci]
EdwardIrby Apr 6, 2026
c544883
refactor: get rid of bootstrpa and rename
EdwardIrby Apr 6, 2026
c7f8d87
feat(behavioral): add deadlock snapshot emission
EdwardIrby Apr 6, 2026
00a0233
feat(behavioral): add structured match listeners
EdwardIrby Apr 6, 2026
e14c78f
feat(behavioral): add verifier-safe listener authoring helpers
EdwardIrby Apr 6, 2026
ec247c2
feat(behavioral): add replay-safe authoring helpers
EdwardIrby Apr 6, 2026
a5f4c6d
feat(behavioral): add dynamic bThreads.spawn API
EdwardIrby Apr 6, 2026
46b329a
feat(behavioral): improve spawned thread snapshot attribution
EdwardIrby Apr 6, 2026
a83bfa6
refactor(behavioral): normalize snapshot thread attribution
EdwardIrby Apr 6, 2026
a567341
refactor(behavioral): extract frontier computation from selectNextEvent
EdwardIrby Apr 6, 2026
76abff9
feat(behavioral): add replay-to-frontier seam
EdwardIrby Apr 6, 2026
f3207c6
feat(behavioral): add replay-safe frontier explorer helper
EdwardIrby Apr 6, 2026
c39f620
test(behavioral): clarify explorer truncated semantics
EdwardIrby Apr 6, 2026
acf12b4
feat(behavioral): add frontier verification helper
EdwardIrby Apr 6, 2026
cba032b
refactor: remove restricted trigger pathway
EdwardIrby Apr 6, 2026
877131d
feat(behavioral): add provenance-aware match listeners
EdwardIrby Apr 6, 2026
e3731bb
feat(agent): align ingress with emit provenance and context memory
EdwardIrby Apr 7, 2026
2d1f12d
feat(agent): publish module install diagnostics to snapshots
EdwardIrby Apr 7, 2026
a87b802
refactor(behavioral): tighten canonical authoring surface
EdwardIrby Apr 8, 2026
923c40a
refactor(behavioral): finalize canonical API and migrate behavioral t…
EdwardIrby Apr 8, 2026
5e8685b
refactor(ui,agent): migrate behavioral contract consumers
EdwardIrby Apr 8, 2026
ca5df74
refactor(agent,behavioral): harden listener selection and module name…
EdwardIrby Apr 12, 2026
79799ed
refactor(agent,behavioral): tighten listener-first module contracts
EdwardIrby Apr 12, 2026
577b304
refactor(agent): enforce local handler normalization in useModule
EdwardIrby Apr 12, 2026
9462944
--wip-- [skip ci]
EdwardIrby Apr 13, 2026
2509ddb
refactor(behavioral): tighten extension listener scoping
EdwardIrby Apr 13, 2026
c095dbe
refactor(behavioral): rename installer module and move installer types
EdwardIrby Apr 13, 2026
a76e457
refactor(behavioral): align installer and frontier tests
EdwardIrby Apr 13, 2026
8209f11
refactor(behavioral): inline trigger ingress and rename prefix constant
EdwardIrby Apr 13, 2026
f36b931
refactor(behavioral): export installer and extension helpers
EdwardIrby Apr 13, 2026
d51da7d
--wip-- [skip ci]
EdwardIrby Apr 13, 2026
bce75ae
refactor(agent): prune stale core type/schema surface
EdwardIrby Apr 13, 2026
b37d000
docs(agent): clarify create-agent core ownership comment
EdwardIrby Apr 13, 2026
c092e50
fix(mcp): correct package metadata import
EdwardIrby Apr 13, 2026
58a1b04
fix(agent): harden core tool-bash requestId gate
EdwardIrby Apr 13, 2026
cd9a93a
feat(agent): add tool-bash request normalizer
EdwardIrby Apr 13, 2026
21922c4
feat(agent): route tool bash execution results
EdwardIrby Apr 13, 2026
2da7b30
feat(modules): add useExtension server transport module
EdwardIrby Apr 13, 2026
2ef347b
chore: dirty refactor
EdwardIrby Apr 13, 2026
c858be3
refactor(ui): inline document runtime core extension
EdwardIrby Apr 13, 2026
5f99d5f
fix(behavioral,server): localize extension events with private bypass…
EdwardIrby Apr 13, 2026
9e1c93a
refactor: remove archived modules and align runtime test surfaces
EdwardIrby Apr 13, 2026
9cdf05a
test: harden server-backed suite stability
EdwardIrby Apr 14, 2026
ab3c4f9
chore(test): add test runner script and align lock/config
EdwardIrby Apr 14, 2026
f5c1a94
docs(skills): add plaited-development workflow skill
EdwardIrby Apr 14, 2026
ce470e8
docs(github): add pr template and description lint
EdwardIrby Apr 14, 2026
c580d3d
docs(skills): add Cline docs MCP skill
EdwardIrby Apr 14, 2026
9ef07ed
chore(dev): add repo-local cline and kanban tooling
EdwardIrby Apr 14, 2026
d6c310f
chore(gitignore): ignore local hyprotherticcal architecture doc
EdwardIrby Apr 14, 2026
1cbf327
chore(dev): align repo-local cline/kanban workflow surfaces
EdwardIrby Apr 14, 2026
da64804
docs(skills): update dev/main workflow and kanban templates
EdwardIrby Apr 14, 2026
91ccce2
chore(workflows): remove claude code review workflow
EdwardIrby Apr 14, 2026
08796b3
chore(workflows): add release-readiness security context
EdwardIrby Apr 14, 2026
3d96e6c
fix(workflows): tolerate unavailable security context (#238)
EdwardIrby Apr 14, 2026
56a7b0b
docs(workflow): clarify agent worktree placement (#239)
EdwardIrby Apr 14, 2026
99ee29e
fix(cli): preserve markdown parsing while avoiding redos (#240)
EdwardIrby Apr 14, 2026
d72bf5f
fix(workflows): harden release-readiness issue generation (#242)
EdwardIrby Apr 14, 2026
6d905a2
chore(workflows): gate trusted review and issue ingress (#243)
EdwardIrby Apr 15, 2026
a8f09fa
docs(skills): document PR branch freshness policy (#244)
EdwardIrby Apr 15, 2026
411fc30
fix(workflows): use release readiness security token (#245)
EdwardIrby Apr 15, 2026
82370da
fix(workflows): stabilize release readiness check sampling (#246)
EdwardIrby Apr 15, 2026
35f5159
feat(workflows): add human-triggered release PR workflow (#247)
EdwardIrby Apr 15, 2026
b7921c3
fix(workflows): ignore release operator checks in readiness (#248)
EdwardIrby Apr 15, 2026
f1de21e
fix(workflows): pin setup-bun action in cline review
EdwardIrby Apr 15, 2026
6ad32fd
fix(workflows): scope external intake to fork PRs
EdwardIrby Apr 15, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
1 change: 1 addition & 0 deletions .agents/skills/add-mcp
1 change: 1 addition & 0 deletions .agents/skills/add-remote-mcp
1 change: 1 addition & 0 deletions .agents/skills/autoresearch-workflows
1 change: 1 addition & 0 deletions .agents/skills/behavioral-core
1 change: 1 addition & 0 deletions .agents/skills/bootstrap-plaited-agent
141 changes: 141 additions & 0 deletions .agents/skills/cmux-browser/SKILL.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,141 @@
---
name: cmux-browser
description: End-user browser automation with cmux. Use when you need to open sites, interact with pages, wait for state changes, and extract data from cmux browser surfaces.
---

# Browser Automation with cmux

Use this skill for browser tasks inside cmux webviews.

## Core Workflow

1. Open or target a browser surface.
2. Verify navigation with `get url` before waiting or snapshotting.
3. Snapshot (`--interactive`) to get fresh element refs.
4. Act with refs (`click`, `fill`, `type`, `select`, `press`).
5. Wait for state changes.
6. Re-snapshot after DOM/navigation changes.

```bash
cmux --json browser open https://example.com
# use returned surface ref, for example: surface:7

cmux browser surface:7 get url
cmux browser surface:7 wait --load-state complete --timeout-ms 15000
cmux browser surface:7 snapshot --interactive
cmux browser surface:7 fill e1 "hello"
cmux --json browser surface:7 click e2 --snapshot-after
cmux browser surface:7 snapshot --interactive
```

## Surface Targeting

```bash
# identify current context
cmux identify --json

# open routed to a specific topology target
cmux browser open https://example.com --workspace workspace:2 --window window:1 --json
```

Notes:
- CLI output defaults to short refs (`surface:N`, `pane:N`, `workspace:N`, `window:N`).
- UUIDs are still accepted on input; only request UUID output when needed (`--id-format uuids|both`).
- Keep using one `surface:N` per task unless you intentionally switch.

## Wait Support

cmux supports wait patterns similar to agent-browser:

```bash
cmux browser <surface> wait --selector "#ready" --timeout-ms 10000
cmux browser <surface> wait --text "Success" --timeout-ms 10000
cmux browser <surface> wait --url-contains "/dashboard" --timeout-ms 10000
cmux browser <surface> wait --load-state complete --timeout-ms 15000
cmux browser <surface> wait --function "document.readyState === 'complete'" --timeout-ms 10000
```

## Common Flows

### Form Submit

```bash
cmux --json browser open https://example.com/signup
cmux browser surface:7 get url
cmux browser surface:7 wait --load-state complete --timeout-ms 15000
cmux browser surface:7 snapshot --interactive
cmux browser surface:7 fill e1 "Jane Doe"
cmux browser surface:7 fill e2 "jane@example.com"
cmux --json browser surface:7 click e3 --snapshot-after
cmux browser surface:7 wait --url-contains "/welcome" --timeout-ms 15000
cmux browser surface:7 snapshot --interactive
```

### Clear an Input

```bash
cmux browser surface:7 fill e11 "" --snapshot-after --json
cmux browser surface:7 get value e11 --json
```

### Stable Agent Loop (Recommended)

```bash
# navigate -> verify -> wait -> snapshot -> action -> snapshot
cmux browser surface:7 get url
cmux browser surface:7 wait --load-state complete --timeout-ms 15000
cmux browser surface:7 snapshot --interactive
cmux --json browser surface:7 click e5 --snapshot-after
cmux browser surface:7 snapshot --interactive
```

If `get url` is empty or `about:blank`, navigate first instead of waiting on load state.

## Deep-Dive References

| Reference | When to Use |
|-----------|-------------|
| [references/commands.md](references/commands.md) | Full browser command mapping and quick syntax |
| [references/snapshot-refs.md](references/snapshot-refs.md) | Ref lifecycle and stale-ref troubleshooting |
| [references/authentication.md](references/authentication.md) | Login/OAuth/2FA patterns and state save/load |
| [references/authentication.md#saving-authentication-state](references/authentication.md#saving-authentication-state) | Save authenticated state right after login |
| [references/session-management.md](references/session-management.md) | Multi-surface isolation and state persistence patterns |
| [references/video-recording.md](references/video-recording.md) | Current recording status and practical alternatives |
| [references/proxy-support.md](references/proxy-support.md) | Proxy behavior in WKWebView and workarounds |

## Ready-to-Use Templates

| Template | Description |
|----------|-------------|
| [templates/form-automation.sh](templates/form-automation.sh) | Snapshot/ref form fill loop |
| [templates/authenticated-session.sh](templates/authenticated-session.sh) | Login once, save/load state |
| [templates/capture-workflow.sh](templates/capture-workflow.sh) | Navigate + capture snapshots/screenshots |

## Limits (WKWebView)

These commands currently return `not_supported` because they rely on Chrome/CDP-only APIs not exposed by WKWebView:
- viewport emulation
- offline emulation
- trace/screencast recording
- network route interception/mocking
- low-level raw input injection

Use supported high-level commands (`click`, `fill`, `press`, `scroll`, `wait`, `snapshot`) instead.

## Troubleshooting

### `js_error` on `snapshot --interactive` or `eval`

Some complex pages can reject or break the JavaScript used for rich snapshots and ad-hoc evaluation.

Recovery steps:

```bash
cmux browser surface:7 get url
cmux browser surface:7 get text body
cmux browser surface:7 get html body
```

- Use `get url` first so you know whether the page actually navigated.
- Fall back to `get text body` or `get html body` when `snapshot --interactive` or `eval` returns `js_error`.
- If the page is still failing, navigate to a simpler intermediate page, then retry the task from there.
4 changes: 4 additions & 0 deletions .agents/skills/cmux-browser/agents/openai.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
interface:
display_name: "cmux Browser"
short_description: "Automate cmux webview surfaces with snapshot/ref workflows."
default_prompt: "Use this skill for browser automation via cmux CLI: identify target surface, snapshot interactive refs, perform actions, wait for state changes, and re-snapshot to avoid stale refs."
122 changes: 122 additions & 0 deletions .agents/skills/cmux-browser/references/authentication.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,122 @@
# Authentication Patterns

Login flows, session persistence, OAuth, and 2FA patterns for cmux browser surfaces.

**Related**: [session-management.md](session-management.md), [SKILL.md](../SKILL.md)

## Contents

- [Basic Login Flow](#basic-login-flow)
- [Saving Authentication State](#saving-authentication-state)
- [Restoring Authentication](#restoring-authentication)
- [OAuth / SSO Flows](#oauth--sso-flows)
- [Two-Factor Authentication](#two-factor-authentication)
- [Cookie-Based Auth](#cookie-based-auth)
- [Token Refresh Handling](#token-refresh-handling)
- [Security Best Practices](#security-best-practices)

## Basic Login Flow

```bash
cmux browser open https://app.example.com/login --json
cmux browser surface:7 wait --load-state complete --timeout-ms 15000

cmux browser surface:7 snapshot --interactive
# [ref=e1] email, [ref=e2] password, [ref=e3] submit

cmux browser surface:7 fill e1 "user@example.com"
cmux browser surface:7 fill e2 "$APP_PASSWORD"
cmux browser surface:7 click e3 --snapshot-after --json
cmux browser surface:7 wait --url-contains "/dashboard" --timeout-ms 20000
```

## Saving Authentication State

After logging in, save state for reuse:

```bash
cmux browser surface:7 state save ./auth-state.json
```

State includes cookies, localStorage, sessionStorage, and open tab metadata for that surface.

## Restoring Authentication

```bash
cmux browser open https://app.example.com --json
cmux browser surface:8 state load ./auth-state.json
cmux browser surface:8 goto https://app.example.com/dashboard
cmux browser surface:8 snapshot --interactive
```

## OAuth / SSO Flows

```bash
cmux browser open https://app.example.com/auth/google --json
cmux browser surface:7 wait --url-contains "accounts.google.com" --timeout-ms 30000
cmux browser surface:7 snapshot --interactive

cmux browser surface:7 fill e1 "user@gmail.com"
cmux browser surface:7 click e2 --snapshot-after --json

cmux browser surface:7 wait --url-contains "app.example.com" --timeout-ms 45000
cmux browser surface:7 state save ./oauth-state.json
```

## Two-Factor Authentication

```bash
cmux browser open https://app.example.com/login --json
cmux browser surface:7 snapshot --interactive
cmux browser surface:7 fill e1 "user@example.com"
cmux browser surface:7 fill e2 "$APP_PASSWORD"
cmux browser surface:7 click e3

# complete 2FA manually in the webview, then:
cmux browser surface:7 wait --url-contains "/dashboard" --timeout-ms 120000
cmux browser surface:7 state save ./2fa-state.json
```

## Cookie-Based Auth

```bash
cmux browser surface:7 cookies set session_token "abc123xyz"
cmux browser surface:7 goto https://app.example.com/dashboard
```

## Token Refresh Handling

```bash
#!/usr/bin/env bash
set -euo pipefail

STATE_FILE="./auth-state.json"
SURFACE="surface:7"

if [ -f "$STATE_FILE" ]; then
cmux browser "$SURFACE" state load "$STATE_FILE"
fi

cmux browser "$SURFACE" goto https://app.example.com/dashboard
URL=$(cmux browser "$SURFACE" get url)

if printf '%s' "$URL" | grep -q '/login'; then
cmux browser "$SURFACE" snapshot --interactive
cmux browser "$SURFACE" fill e1 "$APP_USERNAME"
cmux browser "$SURFACE" fill e2 "$APP_PASSWORD"
cmux browser "$SURFACE" click e3
cmux browser "$SURFACE" wait --url-contains "/dashboard" --timeout-ms 20000
cmux browser "$SURFACE" state save "$STATE_FILE"
fi
```

## Security Best Practices

1. Never commit state files (they include auth tokens).
2. Use environment variables for credentials.
3. Clear state/cookies after sensitive tasks:

```bash
cmux browser surface:7 cookies clear
rm -f ./auth-state.json
```
108 changes: 108 additions & 0 deletions .agents/skills/cmux-browser/references/commands.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,108 @@
# Command Reference (cmux Browser)

This maps common `agent-browser` usage to `cmux browser` usage.

## Direct Equivalents

- `agent-browser open <url>` -> `cmux browser open <url>`
- `agent-browser goto|navigate <url>` -> `cmux browser <surface> goto|navigate <url>`
- `agent-browser snapshot -i` -> `cmux browser <surface> snapshot --interactive`
- `agent-browser click <ref>` -> `cmux browser <surface> click <ref>`
- `agent-browser fill <ref> <text>` -> `cmux browser <surface> fill <ref> <text>`
- `agent-browser type <ref> <text>` -> `cmux browser <surface> type <ref> <text>`
- `agent-browser select <ref> <value>` -> `cmux browser <surface> select <ref> <value>`
- `agent-browser get text <ref>` -> `cmux browser <surface> get text <ref-or-selector>`
- `agent-browser get url` -> `cmux browser <surface> get url`
- `agent-browser get title` -> `cmux browser <surface> get title`

## Core Command Groups

### Navigation

```bash
cmux browser open <url> # opens in caller's workspace (uses CMUX_WORKSPACE_ID)
cmux browser open <url> --workspace <id|ref> # opens in a specific workspace
cmux browser <surface> goto <url>
cmux browser <surface> back|forward|reload
cmux browser <surface> get url|title
```

> **Workspace context:** `browser open` targets the workspace of the terminal where the command is run (via `CMUX_WORKSPACE_ID`), even if a different workspace is currently focused. Use `--workspace` to override.

### Snapshot and Inspection

```bash
cmux browser <surface> snapshot --interactive
cmux browser <surface> snapshot --interactive --compact --max-depth 3
cmux browser <surface> get text body
cmux browser <surface> get html body
cmux browser <surface> get value "#email"
cmux browser <surface> get attr "#email" --attr placeholder
cmux browser <surface> get count ".row"
cmux browser <surface> get box "#submit"
cmux browser <surface> get styles "#submit" --property color
cmux browser <surface> eval '<js>'
```

### Interaction

```bash
cmux browser <surface> click|dblclick|hover|focus <selector-or-ref>
cmux browser <surface> fill <selector-or-ref> [text] # empty text clears
cmux browser <surface> type <selector-or-ref> <text>
cmux browser <surface> press|keydown|keyup <key>
cmux browser <surface> select <selector-or-ref> <value>
cmux browser <surface> check|uncheck <selector-or-ref>
cmux browser <surface> scroll [--selector <css>] [--dx <n>] [--dy <n>]
```

### Wait

```bash
cmux browser <surface> wait --selector "#ready" --timeout-ms 10000
cmux browser <surface> wait --text "Done" --timeout-ms 10000
cmux browser <surface> wait --url-contains "/dashboard" --timeout-ms 10000
cmux browser <surface> wait --load-state complete --timeout-ms 15000
cmux browser <surface> wait --function "document.readyState === 'complete'" --timeout-ms 10000
```

### Session/State

```bash
cmux browser <surface> cookies get|set|clear ...
cmux browser <surface> storage local|session get|set|clear ...
cmux browser <surface> tab list|new|switch|close ...
cmux browser <surface> state save|load <path>
```

### Diagnostics

```bash
cmux browser <surface> console list|clear
cmux browser <surface> errors list|clear
cmux browser <surface> highlight <selector>
cmux browser <surface> screenshot
cmux browser <surface> download wait --timeout-ms 10000
```

## Agent Reliability Tips

- Use `--snapshot-after` on mutating actions to return a fresh post-action snapshot.
- Re-snapshot after navigation, modal open/close, or major DOM changes.
- Prefer short handles in outputs by default (`surface:N`, `pane:N`, `workspace:N`, `window:N`).
- Use `--id-format both` only when a UUID must be logged/exported.

## Known WKWebView Gaps (`not_supported`)

- `browser.viewport.set`
- `browser.geolocation.set`
- `browser.offline.set`
- `browser.trace.start|stop`
- `browser.network.route|unroute|requests`
- `browser.screencast.start|stop`
- `browser.input_mouse|input_keyboard|input_touch`

See also:
- [snapshot-refs.md](snapshot-refs.md)
- [authentication.md](authentication.md)
- [session-management.md](session-management.md)
Loading
Loading