- remove bundled snapshot dependency on src/generated in model-core
- make shared harness provide runtime bundled snapshot
- update guardrail and capability tests to pass explicit snapshot
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
- P1: When allowedEnvVars is provided, PATH was missing from the base
restricted env, causing non-builtin commands to fail at exec time
- P2: Allowlisted HOME/CLAUDE_PROJECT_DIR could overwrite normalized
values from getHomeDirectory()/cwd with ambient process.env values
- P2: Test suite mutated shared pluginHooksState singleton without
resetting it in afterEach, causing cross-test state leaks
- Apply mcp_env_allowlist to plugin hooks: intersect HTTP allowedEnvVars
with MCP allowlist, set command allowedEnvVars to full MCP allowlist
- Scrub process.env in executeHookCommand when allowedEnvVars provided
- Add PluginHooksState class with per-directory Map storage
- Add PluginHooksConfig interface for typed boundary layer
- Pass directory context through hook-config-handler
- Add 16 tests across 4 files (40 assertions) covering allowlist
filtering, env scrubbing, directory isolation, and edge cases
- Remove unnecessary 'as' type assertions, use discriminated union
narrowing instead
OpenAI streaming responses can surface mid-stream errors with type
'server_error' or with the prose message 'An error occurred while processing
your request'. Neither matched any entry in RETRYABLE_MESSAGE_PATTERNS, so
shouldRetryError returned false and the runtime-fallback / fallback-retry
code paths skipped retry. The result was that GPT-5.5 subagent (and main)
turns silently stalled until the stale timeout fired.
The maintainer's diagnosis on issue #3799 explicitly recommends adding
these two patterns to model-error-classifier.ts; this commit does exactly
that and adds two regression tests covering the JSON envelope and the
prose form.
Grok model IDs (e.g. `grok-4.3`, `grok-3-mini`) were missing from `HEURISTIC_MODEL_FAMILY_REGISTRY`, so `resolveCompatibleModelSettings` returned an "unknown family" result for them.
The `chat.params` hook treats unknown families as "no reasoning support" and deletes `output.options.reasoningEffort` before the request leaves the plugin, so users routing Grok via OpenAI-compatible endpoints never saw their reasoning effort forwarded — even with `forceReasoning: true`.
Add a `grok` family entry that includes any model ID containing "grok", exposes `low | medium | high` variants, and sets `reasoningEfforts` so the heuristic surfaces the same capability flag used by the `chat.params` hook. Extend the family table-driven test in `model-settings-compatibility.test.ts` to cover `grok-4.3`.
After the 4.2.0 unified-dispatch refactor (a42f894f / df198d8b / fee515c5 / 989ab717 / dd3fecaf / 1bbe065c / 12bd6580), at least one caller in the new prompt-async-gate path forwards a FallbackModelObject (or some other non-string shape) into parsers that statically claim 'model: string'. The downstream .trim() call then throws 'model.trim is not a function', which rejects the session.processor promise and surfaces as 'Aborted process' + UI 'interrupted'. The issue (#4145) reports this aborts 90% of subagent dispatches across every provider on 4.2.0 + opencode 1.15.4.
This patch adds a 'typeof x !== "string"' runtime guard at the four parser entrypoints called from the dispatch path:
- src/shared/fallback-chain-from-models.ts :: parseVariantFromModel, parseFallbackModelEntry
- src/tools/delegate-task/model-string-parser.ts :: parseVariantFromModelID, parseModelString
- src/shared/model-string-parser.ts (duplicate file with same API) :: parseVariantFromModelID, parseModelString
- src/features/claude-code-agent-loader/claude-model-mapper.ts :: mapClaudeModelString
Each parser now returns undefined / { modelID: "" } for non-string input instead of throwing. This unblocks subagent dispatch and leaves the underlying caller bug for a follow-up.
Regression coverage: three new tests in src/shared/fallback-chain-from-models.test.ts pin the non-string behavior (object, null/undefined, number). Existing 38 tests still pass. Total: 41/41 green, typecheck clean.
Converts the host OpenCode config.skills object (with paths/urls arrays set by other plugins like superpowers) into the SkillsConfig format used by discoverConfigSourceSkills. Filters blank/whitespace entries and non-string values.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
The previous code read params.config.default_agent which is always populated to
"Sisyphus - Ultraworker" by existing logic (lines 196-204) even when the user
never configured default_agent. This silently overrode any custom agent_order.
Now we gate on configuredDefaultAgent (the user's explicit value) so the sort
shim rank map is only mutated when the user actually set default_agent.
Add regression test: agent_order without default_agent must preserve order.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Anthropic made the 1M context window GA for Opus 4.6, Sonnet 4.6, and 4.7 models on March 13, 2026 — no beta header required. The resolver was still falling back to 200K when no modelContextLimitsCache entry existed, causing premature compaction and over-truncation.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
YeonGyu-Kim