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 commands/chappygo-os/atomic-spec/tasksgit clone --depth 1 https://github.com/Chappygo-OS/Atomic-SpecWhat 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 | $0.00021 | $0.09972 |
| Opus 5 | $0.00010 | $0.04986 |
| Sonnet 5 | $0.00004 | $0.01994 |
| Haiku 4.5 | $0.00002 | $0.00997 |
Grade C, and why
tasks scanned grade C with 2 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 2d ago.
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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- Cross-Target Dependency Auto-Generation Rules: When generating tasks, automatically create dependency edges for these patterns: 1. API Producer/Consumer: If a backend task produces an endpoint (e.g., T-020 creates P Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s http://localhost:8000/api/feature/health | jq '.status == "ok"' How it starts
The opening of the file, as written. The whole thing — 1,035 lines — stays where its author put it; the contents beside it link to each section on GitHub.
User Input
$ARGUMENTS
You MUST consider the user input before proceeding (if not empty).
⛔ MANDATORY STRUCTURAL REQUIREMENTS
STOP. Read this section before generating ANY output.
Per Constitution Article IX, this command has NON-NEGOTIABLE constraints:
| Constraint | Required | Forbidden |
|---|---|---|
| Task storage | tasks/ directory with T-XXX-[name].md files |
Single tasks.md file |
| Dashboard | index.md in FEATURE_DIR |
No navigation file |
| Traceability | traceability.md mapping requirements ↔ tasks |
Unmapped tasks |
| Task content | All 4 elements (ID, Mapping, Detail, Verification) | Incomplete tasks |
If you are about to create a file called tasks.md, STOP. You are violating the Constitution.
Outline
1. Setup & Gate Compliance Check
Run {SCRIPT} from repo root. This script will:
- Parse FEATURE_DIR and AVAILABLE_DOCS list
- Automatically validate gate criteria (spec.md, plan.md, HITL approval)
- BLOCK execution if gates fail - you will see error output
If the script outputs gate failures, report them to the user and DO NOT PROCEED.
1.25 Phase 1 Prelude — Efficiency Hint (v0.4+) (advisory)
Run: atomicspec select-model --phase coordinator
If a non-empty model name is printed, the project has configured an advisory tier for coordinator-role turns like task generation; agents that support per-turn model selection (Claude Code subagent Task tool) MAY honor it. If the command prints nothing, advisor_enabled is false (v0.4 default) — proceed with default behavior; byte-for-byte v0.3 compatibility is preserved.
Advisory only: does not affect Directive 3 pinning (task generation is not yet under Context Pinning; that begins at /atomicspec.implement), gate criteria, atomic task naming (Directive 2), or embedded-context generation (Directive 8). Per-feature measurement lands in v0.4.1.
1.5 Load Project Defaults Registry
Per Constitution Article IX, Directive 7 - Load registry before generating tasks.
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.
- 2d ago First seen · 1,035 lines · 21 tokens per session scan C 94c04fba9870
tasks is a command published in the GitHub repository Chappygo-OS/Atomic-Spec (10 stars, last pushed 9d ago), licensed MIT. It adds 21 tokens to every session and 9,972 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 2 findings (hidden instructions, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
constitution
根据交互输入或已提供原则创建或更新项目章程，并确保相关模板保持同步。.
cross-research
Two-phase multi-model research. Phase A fans codex+gemini in parallel (benchmarks + citations angles); Phase B synthesises via fresh Claude session. Usage: /cross-research [--with | list | compare].
ijfw-audit
Run the IJFW audit gate for the current workflow phase. Usage: /ijfw-audit [phase name].
memory-why
Show why a memory recall returned what it did -- BM25 vs vector vs hybrid provenance.
ijfw-execute
Jump directly to the IJFW workflow Execute phase (Deep D4 / Quick Q3). Usage: /ijfw-execute [task or phase name].
ijfw-metrics
Show IJFW token / cost / session metrics over a time window. Reads .ijfw/metrics/sessions.jsonl via the ijfwmetrics MCP tool.