[internal-agent] Two fixed system prompts live outside the shared built-in prompt source and its 3000-code-point test
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 85/100
- Issue type
- Bug
- Clarity
- Clearly specified
- Activity status
- Active
- Tech stack
- typescript
Research direction
Read the spec at docs/specs/20260813-internal_agent_runtime.md and the decision in docs/decisions/20260924-centralized_bounded_builtin_prompts.md. Move POLICY_AGENT_GATE_PROMPT from apps/nanocore/src/policy/agent-gate.ts:16 and COMPACTION_INSTRUCTION from apps/nanocore/src/internal-agents/compacting-context.ts:25 into apps/nanocore/src/internal-agents/builtin-prompts.ts. Done when both texts live there and the existing test builtin-prompts.test.ts:23-30 enumerates them.
Written by the indexing model from the issue text.
Description
Found by the independent documentation audit (supplement on branch audit/independent-docs-20261025, Internal Agent Runtime specification).
- Spec clause:
docs/specs/20260813-internal_agent_runtime.md:251@547887980b6fd3ef730ecc669d39a3092f731408, section "Prompt Contract": "All NanoCore built-in agents keep their fixed System Prompt text in one source location, including callers outside the shared loop and internal model-based helpers." and "A test enumerates all built-in prompt assemblies and enforces this limit." - Decision:
docs/decisions/20260924-centralized_bounded_builtin_prompts.md: "Manage the fixed System Prompt text of every NanoCore built-in agent in one source location." - Code at
547887980:apps/nanocore/src/policy/agent-gate.ts:16definesPOLICY_AGENT_GATE_PROMPTand passes it assystemPromptat line 64.apps/nanocore/src/internal-agents/compacting-context.ts:25definesCOMPACTION_INSTRUCTIONand passes it assystemPromptat line 189.- Neither is in
apps/nanocore/src/internal-agents/builtin-prompts.ts, andbuiltin-prompts.test.ts:23-30checks only the assemblies in that file.
Observed versus required
Required: every fixed built-in prompt, including model-based helpers, lives in one source and is covered by the 3000-code-point test. Observed: the Policy agent gate and the compaction helper keep their own fixed prompts, and no test bounds them.
Adjudication
The code differs from the current accepted specification and decision. Classification: implementation defect.
Repair direction
Move both fixed texts into the shared built-in prompt source, and let the existing test enumerate them.
Severity hint
Low: maintainability and prompt-size control, no data or authorization effect.
- Dominant language
- TypeScript
- Stars
- 0
- Forks
- 1
- Avg merge
- 3h 16m
- Merged PRs (30d)
- 6
Getting set up
- No Dockerfile or Docker Compose file
- No pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from lingkaix/openkit
-
bug doc-audit
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
Maintainers usually reply within 1 day
-
bug doc-audit
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
Maintainers usually reply within 1 day
-
bug doc-audit
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
Maintainers usually reply within 1 day
-
bug doc-audit
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
lingkaix/openkit#491 · 1 comment ·
Maintainers usually reply within 1 day
-
bug doc-audit
Difficulty 2/5 1-3 hours Newbie friendliness 80/100
Maintainers usually reply within 1 day
All issues in lingkaix/openkit
Similar issues
-
area:plugins area:security area:skills bug has repro
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
anthropics/claude-code#105911 ·
Maintainers usually reply within 1 day
-
openspec-apply-change skill: when blocked, may select a ready artifact unrelated to missingArtifactsOpen
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Fission-AI/OpenSpec#2165 ·
Maintainers usually reply within 1 day
-
🐞 bug 🧐 unconfirmed
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
Tencent/tdesign-vue-next#7032 · 1 comment ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
gnosis/gnosis_vpn#661 ·
Maintainers usually reply within 1 day
-
content-freshness: staleOwnedContent doc comments still say 'complete receipt' (follow-up to #7912)Opentriage:bug-verified
Difficulty 1/5 Under an hour Newbie friendliness 92/100
Maintainers usually reply within 1 day