Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/loki-ai-ch/spec-manager/codebuddygit clone --depth 1 https://github.com/loki-ai-ch/spec-managerWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00524 |
| Opus 5 | $0.00000 | $0.00262 |
| Sonnet 5 | $0.00000 | $0.00105 |
| Haiku 4.5 | $0.00000 | $0.00052 |
Grade A, and why
CODEBUDDY scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
88% identical to CLAUDE — 16 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 28 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spec-Driven Development
This project uses spec-manager for local-first spec-driven development. CodeBuddy should use the project skill at .codebuddy/skills/spec-manager/ when the user asks for /spec-manager or asks to follow the spec-manager workflow.
Unified Rules
- Feature work MUST go through
spec-manager. - New or non-trivial work follows L1 -> L2 -> L3 -> Agent Task.
- Never write implementation code without a frozen L3 spec.
- L1/L2 approval advances
draft -> confirmed; one explicit L3 approval (an explicit user approval) advancesdraft -> frozen. - Before creating a new spec, inspect existing specs and decisions.
- Before code edits, read the frozen L3 spec and create/start an Agent Task.
- planJson
coveredSpecsMUST include the current L3 specCode. - If adaptive workflow is enabled, Task creation records a
standardorgovernedProfile snapshot;governedrequires the frozen L3 to declare## 关键验收标准with valid AC IDs, and task complete requires successful verification evidence covering every critical AC.standardreports missing coverage as warnings. Usespec-manager project profile recommend --request "<work>"for a deterministic, explainable recommendation; it does not auto-enable adaptive workflow and is not a hidden gate. Usespec-manager project profile metrics [--topic <topic>] [--json]for a read-only governance report over Profile adoption, governed coverage gaps, standard warnings, and explicit overrides; metrics does not modify config or historical Tasks. Usespec-manager project readiness critical [--topic <topic>] [--json]for a read-only critical AC readiness report and repair suggestions; it must not auto-generate or insert critical AC. Before enabling adaptive workflow, usespec-manager project workflow preview [--json]for a read-only adoption preview; preview does not write config, migrate historical Tasks, or act as an enable gate. quickremains a restricted lightweight exception and does not create the full L1/L2/L3/Task chain.- Validate L3 markdown plans with
spec-manager spec validate-plan --from-spec <L3-code>. - Record execution with
spec-manager task step; finish withspec-manager task complete.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 28 lines · 0 tokens per session scan A e3a3c049fbc8
CODEBUDDY is an agent published in the GitHub repository loki-ai-ch/spec-manager (5 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 524 tokens. A static security scan graded it A with 0 findings. It is 88% identical to CLAUDE, differing in 16 lines, and is treated as a copy.
Other agents, from other repositories
planner
Drafts the execution Plan (plan.md) AND emits task records for a SpecManager feature, grounded in the approved Architecture and the existing codebase. Plans MUST be organised into phases with Fibonacci-scored tasks ≤3.
walkthrough-writer
Writes a per-phase Walkthrough document for a SpecManager feature, or — in phase: "final" mode — a feature-level roll-up that links each phase walkthrough.
architect
Drafts an Architecture document for a SpecManager feature, grounded in the approved PRD AND the existing codebase. Reads repo source/conventions before writing.
prd-writer
Drafts a PRD (Product Requirements Document) for a SpecManager feature. Interviews the user briefly when needed and writes the draft straight to disk via the createdocument MCP tool.
reviewer
Read-only spec-compliance reviewer. Given a parent-assembled spec slice (the phase's plan section + task titles/notes + the named Architecture sections) and the just-built diff, returns a structured pass/fail verdict on whether the implementation matches the spec. Never writes. Invoked by /specmanager-build after the…
wcag-audit
Accessibility audit based on WCAG 2.1/2.2. Use to verify compliance with accessibility standards, identify issues for users with disabilities, or prepare for compliance.