The migration log message previously pointed users to docs/reference/configuration.md for the new LSP config location, but that doc section still showed the obsolete plugin-level 'lsp' block. A user following the guidance would re-add the same 'lsp' key, see it stripped again on next startup, and never reach a usable config.\n\nFix both sides: rewrite the log message so it is self-contained (states the new path .opencode/lsp.json and the consumer directly) and rewrite the LSP section in docs/reference/configuration.md to describe the actual current architecture (LSP served by the 'lsp' MCP server, reading server map from .opencode/lsp.json via LSP_TOOLS_MCP_PROJECT_CONFIG, schema lives in packages/lsp-tools-mcp).\n\nVerification: bun test src/shared/migration/ -> 26/26 pass. bun run typecheck -> exit 0. Manual probe -> migration still strips lsp from both in-memory and persisted file.
The hold default jumped 8x in v4.2.3 to absorb slower-provider
session.error arrivals before reservation release. Update the RFC code
block, the prose default callout, and the retry-latency note so the
documented default matches src/shared/prompt-async-gate.ts.
Captures the porting decisions made in this round (pi-rules fingerprinting + omo transcript-aware dedup) along with the pre/post functional and performance gap matrices for codex-rules, pi-rules, and the omo rules-injector.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Refs #3772 (the rotation half — EPIPE shutdown-noise suppression
remains a separate follow-up).
`src/shared/logger.ts` appends every entry to `os.tmpdir()/oh-my-opencode.log`
via `fs.appendFileSync` with no size cap. On long-running or busy projects
the file grows into the multi-GB range — a real-world reproduction on one
machine showed a 4.5 GB `oh-my-opencode.log.1` accumulated from per-shutdown
noise across many sessions. Eats `%TEMP%` on Windows and `/tmp` on Unix.
Add size-based rotation inside the existing batched `flush()` path:
oh-my-opencode.log → oh-my-opencode.log.1
oh-my-opencode.log.1 → oh-my-opencode.log.2 (oldest dropped)
Cap is 50 MB per file; worst-case on-disk footprint is therefore ~150 MB.
The check runs only inside `flush()`, so the cost is amortized over
`BUFFER_SIZE_LIMIT` (50 entries) or the 500 ms flush timer. All filesystem
ops stay wrapped in try/catch — logging must never throw — and a failed
rotation leaves existing on-disk state intact rather than crashing the
agent. Pattern mirrors `src/openclaw/reply-listener-log.ts`, but with two
backup slots instead of one to keep a usable history window for debugging.
No config knobs in this iteration. The issue proposes `logs.max_size_mb`
/ `logs.max_files`, but the defaults are reasonable and adding schema is
more surface area than the bug warrants. Easy to promote later (the
existing test seams already let callers override the cap).
Tests:
- `src/shared/logger.test.ts` (new): under-threshold no-rotate, over-
threshold rotates to `.1`, repeated rotation evicts oldest, rotation-
failure-doesn't-throw, default path lives under `os.tmpdir()`. Uses a
`mock.module(...)` substring marker so `script/run-ci-tests.ts` routes
the file to its own bun process — the logger module's singleton state
otherwise gets contaminated by sibling tests that mock `./shared`.
Out of scope: suppressing specific shutdown-noise messages (EPIPE,
`unhandledRejection received during shutdown cleanup`). The rotation
cap bounds the disk impact regardless of which noise pattern is
generating volume; per-message suppression can stand on its own
merits in a follow-up.
oh-my-openagent injects MCP servers (websearch, context7, grep_app) at
runtime via the OpenCode plugin API. The `opencode mcp list` command reads
only OpenCode's static config and therefore reports no servers even though
the plugin MCPs are active — this is expected, not a bug.
Add a "Native vs plugin-injected MCPs" subsection to docs/reference/features.md
that explains the three-tier architecture, shows the visibility table, and
points users to `bunx oh-my-openagent doctor --verbose` for runtime
inspection. Add brief inline notes in README.md at both MCP bullet points.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Race-condition and concurrency fixes must include reporter-verified repro confirmation before the originating issue is closed. Adds the checklist and rationale grounded in recent incident examples.
Closes MEDIUM-12
Documents the reservation-based duplicate-injection guard introduced in v4.2.0 with accepted status, exported API signatures, release semantics, migration notes, and commit references.
Closes MEDIUM-11
PR #3825 introduced a delegated child-session bootstrap to capture first-prompt retry payloads before history is persisted, addressing the empty-history fallback gap. After merge the PR's own regression test failed on clean root bun test (6828 pass / 1 fail), so PR #4044 reverted it. Ship v4.2.0 with the bug documented and a workaround so users have an explicit story for the unfixed delegated child-session early-failure path. Reland will target v4.2.1.
Closes BLOCKER-4 (Path B - reland deferred to v4.2.1)
PR #3825 added a shared bootstrap context to capture delegated
child-session retry payloads before the first prompt dispatch, so
empty-history failures could still retry through the fallback chain.
The PR's own regression test failed on clean root bun test after merge
(6828 pass / 1 fail). PR #4044 reverted the merge to keep dev green.
Ship v4.2.0 with the bug documented and a workaround so users have an
explicit story for the unfixed delegated child-session early-failure
path. Reland targets v4.2.1 once the regression test is stabilized.
Closes BLOCKER-4 (Path B - documentation, reland deferred to v4.2.1)
Race-condition and concurrency fixes must include reporter-verified repro
confirmation before the originating issue is closed. CI green is necessary
but not sufficient.
Closes M12
PR #3825's fac90d69f introduced a shared bootstrap context to fix delegated child-session fallback when the first prompt fails before any session history is persisted. PR #4044 reverted that fix because its own regression test failed on a clean root suite (6828 pass / 1 fail). The bug remains unaddressed in v4.2.0; reland is deferred.
This commit documents the symptom, history, workaround, and tracking issue so users have visibility.
Closes BLOCKER-4 via Path B (documentation).
Refs PR #3825, PR #4044, issue #4059.
PR #3825 introduced a delegated child-session bootstrap to capture first-prompt retry payloads before history is persisted, addressing the empty-history fallback gap. After merge the PR's own regression test failed on clean root bun test (6828 pass / 1 fail), so PR #4044 reverted it. Ship v4.2.0 with the bug documented and a workaround so users have an explicit story for the unfixed delegated child-session early-failure path. Reland will target v4.2.1.
Closes BLOCKER-4 (Path B - reland deferred to v4.2.1)
Follow-up pass to catch drift my earlier docs commit missed:
docs/reference/features.md:
- 'Architecture Snapshot': 5-tier hook composition '52 base / 59 with team mode'
to '54 base / 61 with team mode' (matches actual ToolGuard +2 from
fsync-skip-warning + bash-file-read-guard).
docs/reference/cli.md:
- Add 'boulder' subcommand to Commands table (inspects boulder work-state).
docs/reference/configuration.md:
- Code comment example 'Kimi K2.5 work best' to 'Kimi K2.6 work best' to match
current primary Sisyphus fallback chain.
docs/guide/installation.md:
- Add 'Kimi K2.6 (opencode-go, vercel)' row to the provider availability table
above the existing K2.5 row.
- Selection priority footnote and 'Safe Overrides' chains updated to mention
K2.6 alongside K2.5 in Sisyphus and Atlas chains.
docs/guide/agent-model-matching.md:
- Provider table 'Kimi K2.5' row to 'Kimi K2.6 / K2.5'.
- Atlas fallback chain mentions K2.6 instead of K2.5 alone.
Align README.md, CONTRIBUTING.md, and docs/ with the same ground truth used
in the hierarchical AGENTS.md regeneration commit 1e7a7600a:
- README.md: '25+ built-in hooks' was undercounted by ~30 hooks; now reads
'54+ lifecycle hooks (61 with Team Mode)' to match actual tier composition.
- CONTRIBUTING.md: src/hooks/ '52 hooks across 55 modules' -> '54 base (61 team) across 58 dirs';
src/tools/ '26 tools across 16 directories' -> '20-39 tools (config-gated)';
src/features/ '19 feature modules' -> '20 feature modules' (boulder-state, team-mode added).
- docs/reference/features.md: Core-agent order indices were off-by-one;
Sisyphus 1/Hephaestus 2/Prometheus 3/Atlas 4 -> 0/1/2/3 (matches installAgentSortShim).
- docs/guide/overview.md: Sisyphus model recommendations now mention Kimi K2.6 alongside K2.5
since K2.6 is the active default fallback in the primary Sisyphus chain
(see src/shared/model-requirements.ts).
The manifesto's project-context block carried the typo `oh-my-openagent.com`.
The actual canonical domain has no dashes:
- DNS: `ohmyopenagent.com` resolves (Cloudflare).
- DNS: `oh-my-openagent.com` does not resolve.
- web/wrangler.toml routes ohmyopenagent.com + ohmyopencode.org.
- web/middleware.ts redirects ohmyopencode.org -> ohmyopenagent.com.
Same fix for the legacy alias (ohmyopencode.org, also no dashes) plus an
explicit note that the legacy host 308-redirects to the canonical one so
nobody pastes the dashed form anywhere.
The project-context block had been sitting as a stale local working-tree
edit from earlier audit work and never landed on dev. Promoting it as a
clean commit so the manifesto carries the same project-context summary
the README does.
In #3859 I bulk-replaced `oh-my-opencode.schema.json` ->
`oh-my-openagent.schema.json` across docs/, on the false assumption
that the schema file had been renamed alongside the package.
It hadn't. Repo state on dev:
- assets/oh-my-opencode.schema.json (only file present)
- package.json export `./schema.json` -> `./dist/oh-my-opencode.schema.json`
- script/build-schema.ts emits the same legacy basename
- root AGENTS.md line 135 documents the same legacy URL
So every doc that pointed at `oh-my-openagent.schema.json` was
returning HTTP 404 for any user copying the example into their config.
Verified live:
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/dev/assets/oh-my-openagent.schema.json -> 404
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/dev/assets/oh-my-opencode.schema.json -> 200
Files reverted to canonical legacy URL:
- docs/examples/{coding,default,planning}-focused.jsonc
- docs/guide/{overview,installation,agent-model-matching}.md
- docs/reference/configuration.md (2 occurrences)
Hephaestus's fallbackChain in src/shared/model-requirements.ts has a
single entry: gpt-5.5. The previous wording said "GPT-5.4/5.5" which
implied either was acceptable; only gpt-5.5 is. Note OpenAI API access
as an alternative to ChatGPT Plus/Pro for clarity.
Audit pass against root AGENTS.md (regenerated 2026-05-08) and the
canonical sources in src/shared/model-requirements.ts,
src/features/team-mode/, src/cli/, src/config/schema/. Each touched file
keeps its existing tone but corrects stale facts:
- guide/overview.md: project name + dual-package transition reality
- guide/installation.md: bun-only invocation, dual oh-my-openagent /
oh-my-opencode CLI surface, current doctor categories
- guide/orchestration.md: 11 agents, category routing through
sisyphus-junior, background concurrency, team-mode link
- guide/agent-model-matching.md: Metis primary -> claude-sonnet-4-6
(matches src/shared/model-requirements.ts fallback chain)
- guide/team-mode.md: full schema, member eligibility tiers, 12 team_*
tools added when enabled
- reference/cli.md: trim to current Commander surface (install, doctor,
run, mcp-oauth, refresh-model-capabilities, get-local-version)
- reference/configuration.md: trim to schema-driven reference, JSONC
rules, multi-level merge order, mcp_env_allowlist user-only constraint
- reference/features.md: 5-tier hook composition counts, 3-tier MCPs,
hashline edit, IntentGate, OpenClaw bidirectional dispatcher
Cross-cutting: every `oh-my-opencode.schema.json` URL replaced with
the canonical `oh-my-openagent.schema.json` (legacy package name in
text bodies preserved where it documents the dual-publish transition).
Source code change:
- src/shared/model-requirements.ts: prepend claude-sonnet-4-6 to metis fallback
chain so Sonnet becomes the default. Opus 4.7 max remains as the immediate
fallback for callers who want extra reasoning.
- src/shared/model-requirements.test.ts: update assertion to expect Sonnet
primary + Opus secondary.
AGENTS.md accuracy fixes (verified against source):
- Agent modes: Sisyphus/Hephaestus are 'primary' (not 'all'); Sisyphus-Junior
is 'subagent' (not 'all'). Confirmed via 'const MODE: AgentMode = ...' in
each agent file. Also clarified Prometheus has no agentSources factory and
is built via buildPrometheusAgentConfig.
- Sisyphus fallback chain: corrected order to kimi-k2.6 → k2p5 → kimi-k2.5
→ gpt-5.5 medium → glm-5 → big-pickle (was missing kimi-k2.5).
- Librarian/Explore: added missing minimax-m2.7 step between -highspeed and
claude-haiku-4-5.
- Metis chain: removed fictitious gemini-3.1-pro entry.
- Sisyphus-Junior chain: spelled out the actual fallback (was 'user-configurable').
- Temperatures: Sisyphus/Hephaestus do not set explicit temperature (model
default); Sisyphus-Junior is 0.1 via SISYPHUS_JUNIOR_DEFAULTS.
- Quick category default: gpt-5.4-mini (not gpt-5.4-mini-fast).
Team-mode corrections:
- Eligibility registry has 3 verdicts: eligible (sisyphus, atlas, sisyphus-junior),
conditional (hephaestus — needs D-36 teammate permission), hard-reject
(oracle, librarian, explore, multimodal-looker, metis, momus, prometheus).
- Schema has 11 fields, not 4: added max_messages_per_run, max_wall_clock_minutes,
max_member_turns, base_dir, message_payload_max_bytes, recipient_unread_max_bytes,
mailbox_poll_interval_ms.
- Hooks: 'team-session-events' is 4 sub-handlers in src/plugin/event.ts
(team-idle-wake-hint, team-lead-orphan-handler, team-member-error-handler,
team-member-status-handler), not a single Continuation-tier hook.
- Tier counts now show base + team-mode: ToolGuard 14/15, Transform 5/7.
- Total: 52 base hooks, 59 with team-mode.
Doc cascade for the Metis change:
- docs/guide/orchestration.md, agent-model-matching.md, installation.md
- docs/reference/configuration.md, features.md
The caller-facing description told orchestrators WHAT `deep` is but
never specified goal cardinality, so they bundled multiple goals into
one call (e.g. "fix X + merge Y + deploy Z"). The deep agent's own
prompt was already configured to refuse such bundles, but only after
they arrived. The rule now lives on the caller side: ONE goal + ONE
deliverable per call; multiple goals must fan out as parallel `deep`
calls.
The documented chains were missing entries and had wrong providers:
- Librarian was missing openai/gpt-5.4-mini-fast, vercel/minimax-m2.7-highspeed,
and the last two entries. Also had opencode/ instead of vercel/ for
minimax-m2.7-highspeed.
- Explore had github-copilot|xai/grok-code-fast-1 as primary (which doesn't
exist in the fallback chain) and was missing openai/gpt-5.4-mini-fast.
Both now match the exact runtime chains from model-requirements.ts.
MoerAI