Replace oracle as the agent for council background tasks with a purpose-built
council-member agent. This avoids coupling to oracle's config/prompt and provides
proper read-only tool restrictions (deny write, edit, task, athena_council).
- New council-member-agent.ts with analysis-oriented system prompt
- Registered in agentSources (hidden from Sisyphus delegation table)
- Added to type system, Zod schemas, display names, tool restrictions
- Minimal model fallback (always overridden per council member at launch)
- Council orchestrator now launches members as council-member agent
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Council members launched as agent='athena' got Athena's system prompt saying
'ALWAYS call athena_council first', plus the tool wasn't denied for bg athena
tasks. Each council member spawned 4 more → exponential explosion (47+ tasks).
Three fixes:
1. Deny athena_council in ATHENA_RESTRICTIONS (agent-tool-restrictions.ts)
- Only affects background athena tasks (task-starter.ts)
- Primary Athena (user-selected) still has access via permission field
2. Session-level dedup guard prevents re-calling while council is running
- If Athena retries during long wait, returns 'already running'
3. Increase wait timeout from 2min to 10min (council members need time
for real code analysis with Read/Grep/LSP)
- Implement executeCouncil with parallel member launch and partial-failure tolerance
- Add result collection mapping and wire Athena exports with read-only athena tool restrictions
Blocking fixes:
- B1: Return empty restrictions for unknown/custom agents instead of
EXPLORATION_AGENT_DENYLIST, allowing custom agents full tool access
- B2: Use Object.create(null) consistently across all 5 agent-loading
result objects to prevent prototype pollution
- B3: Add code comment documenting custom agent bash access trust model
- B4: Mock getOpenCodeConfigDir in opencode-config-agents-reader tests
to prevent global config dir leakage
Non-blocking fixes:
- N1: Use resolveAgentDefinitionPaths with project boundary enforcement
in opencode-config-agents-reader for path containment
- N2: Add session-scoped 30s TTL cache to resolveCallableAgents to
avoid redundant SDK IPC calls per tool invocation
- N3: Extract shared parseToolsConfig into src/shared/parse-tools-config.ts
replacing 4 duplicated local implementations
- N4: Add .min(1) to AgentDefinitionPathSchema rejecting empty paths
- N5: Add resolve-agent-definition-paths.test.ts covering tilde expansion,
relative paths, boundary enforcement, and null containmentDir
- N6: Validate agent mode against allowed values instead of bare type
assertion in opencode-config-agents-reader
Wave 1 of agent definitions enhancement (PR #2299):
Schema & Configuration:
- Add agent_definitions field to oh-my-opencode config schema
- Support list of file paths to .md or .json agent definition files
- Add to PARTIAL_STRING_ARRAY_KEYS for Set-union merge semantics
- Implement eager path resolution in loadPluginConfig() before merging
Path Resolution:
- Create resolve-agent-definition-paths.ts helper
- User-level paths resolve from ~/.config/opencode/ (no containment)
- Project-level paths resolve from project root (with containment check)
- Homedir expansion, absolute/relative path handling
JSON Agent Loader:
- Create parseJsonAgentFile() for .json/.jsonc agent definitions
- Validate required fields (name, prompt)
- Support tools as string (comma-separated) or array
- Map model via mapClaudeModelToOpenCode()
- Comprehensive test suite (7 test cases, all passing)
Type Extensions:
- Extend AgentScope: add 'definition-file' and 'opencode-config'
- Add AgentJsonDefinition interface for JSON agent schema
All automated checks passing:
- lsp_diagnostics clean on all changed files
- json-agent-loader.test.ts: 7/7 passing
- Full typecheck: zero new errors
- QA evidence saved to .sisyphus/evidence/
- #3354: Coerce data.name to String in loadSkillFromPath/loadSkillFromPathAsync
to prevent crash when YAML parses numeric skill names (e.g., name: 12306)
- #3416: Add required run_in_background parameter to all task() examples in
ultrawork prompts (default, gpt, gemini, planner) to match tool schema
- #3379/#3417/#3418/#3337/#3335: Strip ZWSP (U+200B) before agent name
comparisons in agent-tool-restrictions, sync-prompt-sender, tool-execute-after,
tool-execute-before, oracle-verification-detector, call-omo-agent,
recovery-prompt-config, and agent-variant to prevent ZWSP-prefixed display
names from breaking exact-match lookups
- shell-env: detect Git Bash via MSYSTEM env var when SHELL is unset (#3366)
On some Git Bash installations SHELL is not set but MSYSTEM (MINGW64/MSYS)
is always present. Check MSYSTEM before PSModulePath to avoid emitting
PowerShell syntax in bash shells.
- session-state: resolve legacy agent names in resolveRegisteredAgentName (#3272)
Historical sessions stored agent names like 'Sisyphus (Ultraworker)' which
don't match the current registered format. Fall back to getAgentConfigKey
for legacy/parenthesized name resolution before returning the raw name.
- config-migration: skip backup when file content is unchanged (#3222)
Compare serialized config with existing file content before creating a
timestamped .bak file. Only create backup when the on-disk content
actually differs from the migrated content.
Claude version transforms now use a single regex
claude-(\w+)-(\d+)-(\d+) -> claude-$1-$2.$3 instead of one
.replace() per model. New claude models need zero changes.
Adds vercel to every fallback entry where the model exists on the
gateway (kimi-k2.5, glm-5, glm-4.6v, minimax-m2.7, grok-code-fast-1,
gpt-5-nano). Adds inferSubProvider mappings for minimax, moonshotai,
and zai. Updates librarian/explore special cases to prefer minimax
via gateway over claude-haiku.
The gateway uses google/gemini-3-flash (no -preview suffix) unlike the
direct Google API. Give the vercel provider its own transform instead
of delegating to sub-provider transforms blindly.
Adds 'vercel' as a recognized provider throughout the model resolution
system, enabling users with Vercel AI Gateway to use it as a universal
fallback for OpenAI, Anthropic, and Google models.
The gateway transform infers the sub-provider from canonical model names
(claude->anthropic, gpt->openai, gemini->google) and delegates to the
appropriate provider-specific transform, producing model strings like
vercel/anthropic/claude-opus-4.6.
The @scope/path regex incorrectly matched npm scoped package references
like 'require(\"@scope/pkg\")' or '--package=@scope/pkg'. Added context
filtering to exclude matches preceded by npm/import indicators.
🤖 Generated with OhMyOpenCode assistance
https://github.com/code-yeongyu/oh-my-opencode
When model metadata reports maxOutputTokens as 0 or negative, the
clamp logic would set maxTokens to 0 and break generation. Added
a positive-value guard so invalid metadata is ignored and the
user-requested maxTokens is preserved.
🤖 Generated with OhMyOpenCode assistance
https://github.com/code-yeongyu/oh-my-opencode
Anthropic API accepts claude-opus-4.6 (dot format) but not
claude-opus-4-6 (dash format). Added anthropic case to
transformModelForProvider to normalize dash-format model IDs before
they reach the provider. Updated model config snapshots and test
expectations to match the new dot-format output.
🤖 Generated with OhMyOpenCode assistance
https://github.com/code-yeongyu/oh-my-opencode
Old sessions and configs reference agent names in parenthesized format
like 'Sisyphus (Ultraworker)' and 'Atlas (Plan Executor)' which failed
to map during migration. Added entries for all six core agents.
🤖 Generated with OhMyOpenCode assistance
https://github.com/code-yeongyu/oh-my-opencode
On Windows, renameSync fails with EPERM/EACCES when the target file
already exists. Fall back to unlink + rename on Windows permission
errors while preserving atomic semantics on other platforms.
🤖 Generated with OhMyOpenCode assistance
https://github.com/code-yeongyu/oh-my-opencode
Previously returned archivedLegacyConfig which was false when archive
rename failed, even though the canonical file was successfully written.
Archive failure is secondary cleanup and should not be reported as
migration failure.
🤖 Generated with OhMyOpenCode assistance
https://github.com/code-yeongyu/oh-my-opencode
Doctor checks spawned subprocesses (gh, opencode, sg) without timeouts,
causing the process to hang indefinitely if any binary was stuck.
The OS would then SIGKILL the process (exit code 137).
- Add spawnWithTimeout utility with 10s per-spawn timeout and proper
cleanup (timer cleared on success, proc.exited awaited after kill)
- Capture both stdout and stderr to preserve gh auth status behavior
- Add 30s overall doctor command timeout with JSON-mode support
- Distinguish timeout errors from other failures in runner
- Update all doctor check subprocess calls to use timeouts
posthog-node SDK defaults disableGeoip to true, preventing country/city
data from being populated despite $ip being sent. Explicitly set
disableGeoip: false to enable geographic analytics.
On Windows, PSModulePath is always set by the system even when the
active shell is Git Bash (via SHELL env var). This caused the
non-interactive-env hook to generate PowerShell syntax ($env:VAR=val)
for bash shells, resulting in 15+ errors per command.
The fix prioritizes SHELL over PSModulePath since an explicit SHELL
variable indicates the user's chosen shell, while PSModulePath is just
a system-wide module path that doesn't indicate the active shell.
Fixes: #3338
Addresses Cubic review: configs without prior _migrations now get the
full migration set written to config as fallback when sidecar write
fails, preventing migration tracking loss.
Update various modules to use centralized constants from plugin-identity:
- get-local-version/formatter: Use PUBLISHED_PACKAGE_NAME
- run/session-resolver: Use PUBLISHED_PACKAGE_NAME
- background-agent/task-poller: Use PUBLISHED_PACKAGE_NAME
- mcp-oauth/provider: Use PUBLISHED_PACKAGE_NAME
- auto-update-checker/constants: Use ACCEPTED_PACKAGE_NAMES
- comment-checker/downloader: Use PUBLISHED_PACKAGE_NAME
- legacy-plugin-toast/hook: Use PLUGIN_NAME
- shared/data-path: Use CACHE_DIR_NAME
- shared/external-plugin-detector: Use ACCEPTED_PACKAGE_NAMES
- shared/logger: Use LOG_FILENAME
- tools/ast-grep/downloader: Use PUBLISHED_PACKAGE_NAME
- tools/call-omo-agent/tools: Use PUBLISHED_PACKAGE_NAME
- tools/delegate-task/category-resolver: Use PUBLISHED_PACKAGE_NAME
- tools/grep/constants: Use PUBLISHED_PACKAGE_NAME
- tools/grep/downloader: Use PUBLISHED_PACKAGE_NAME
- tools/lsp/lsp-client-wrapper: Use PUBLISHED_PACKAGE_NAME
🤖 Generated with assistance of OhMyOpenCode
Add CONFIG_BASENAME and LEGACY_CONFIG_BASENAME constants to support
migrating from oh-my-opencode.json to oh-my-openagent.json naming.
🤖 Generated with assistance of OhMyOpenCode
The resolveModelPipeline test was consistently failing on CI
(resolveModelPipeline > does not return unused explicit user config
metadata in override result) while passing locally. Root cause is the
same mock.module contamination pattern as #3243: when bun runs all test
files in a single process, mock.module calls from other files (e.g.
model-resolver.test.ts) leak into this file's module scope.
Add mock.module('./logger', ...) so run-ci-tests.ts auto-detects the
file and runs it in its own isolated bun process.
This unblocks the v3.16.1 publish workflow which hit this flake twice.
ismeth