task-pipeline CLAUDE.md

task-pipeline CLAUDE.md is an instructions file for coding agents from SpaiR/task-pipeline. It costs 3,721 tokens per session, scanned A, original, MIT.

Repository guidance for Claude Code, the coding assistant, explaining how to edit this project’s Markdown-based skills and task files.

In plain words
What is it for?
Use it when editing skills, task records, plans, specifications, or other project documentation.
Why use it?
It prevents changes that violate the repository’s documented structure, artifact rules, and commit format.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/spair/task-pipeline/claude-md
Clone the repo
git clone --depth 1 https://github.com/SpaiR/task-pipeline

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for task-pipeline CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/spair/task-pipeline/claude-md.svg)](https://agentmods.dev/instructions/spair/task-pipeline/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/spair/task-pipeline/claude-md"><img src="https://agentmods.dev/badge/instructions/spair/task-pipeline/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 3,721 This file is loaded in full into every session.
When invoked 3,721 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.03721 $0.03721
Opus 5 $0.01861 $0.01861
Sonnet 5 $0.00744 $0.00744
Haiku 4.5 $0.00372 $0.00372

Measured today against content hash 651f2213d61a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

task-pipeline CLAUDE.md 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 today.

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.

CLAUDE.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Guidance for Claude Code when editing this repository. User-facing documentation lives in README.md; the full artifact contract lives in docs/contract.md.

Quick orient

A collection of user-invocable Claude Code skills implementing a chat-first "context serialization protocol", not an orchestration engine: discuss freely in chat, then fix the discussion into a fixed-format Markdown artifact under .task/ with one short skill. Depth of capture is the skill name, never a flag. Skills in skills/<name>/SKILL.md; the plugin ships exactly one agent, agents/code-reviewer.md (the review pass — see the invariants below), and no phase companions, no lock protocol, no hook gate. No build/test/lint. Work here is editing markdown (occasional bash) and reasoning about pipeline semantics.

discuss freely in chat
  ↓
grill                                 ← pre-capture: interrogate the decision, no artifact
  ↓
to-task | to-plan | to-roadmap        ← capture depth is the skill, not a flag
to-spec                               ← pins technical decisions, cited via Spec:
  ↓                       ↓
implement session   roadmap-to-workflow   ← the launcher fans items out to sessions
  ↓                       ↓
task:code-reviewer                    ← the plugin's own review pass, spawned by both:
                                        prove → fix within Touches → Build and Tests → amend

grill sits at the "discuss freely" stage: it interrogates a plan/decision one question at a time, keeps a decision-plus-rationale ledger, ends with a pre-mortem, and routes to the right capture skill — it writes no artifacts and touches nothing under .task/. to-task captures ## Description only into .task/task/<slug>.md; to-plan adds ## Plan (Goal/Touches/Logic steps); to-roadmap captures a multi-task initiative into .task/roadmap/<slug>.md; to-spec captures load-bearing technical decisions into a standalone .task/spec/<slug>.md, referenced by tasks/roadmaps via a Spec: header and read by the executing session as a fixed anchor. There is no execution skill — every artifact carries a stamped one-line ## Execution pointer to .task/CLAUDE.md## Executing a task, and an ordinary session told implement .task/task/<slug>.md follows it: implement the plan, commit per .task/CLAUDE.md → Commit Format, spawn task:code-reviewer on the resulting diff (it proves each finding, fixes what it confirms within Touches, runs .task/CLAUDE.md → Build and Tests, and amends the commit), tick the roadmap item if Roadmap:/Source item: are present. roadmap-to-workflow is the one launcher: it computes dependency-ordered waves over a roadmap's unchecked items and invokes the plugin-shipped Workflow driver (skills/_lib/roadmap-driver.js, via scriptPath + args) — opus-plans (sonnet for haiku-hinted items)/sonnet-implements/task:code-reviewer-reviews per item by default, plus the driver's own mark stage ticking the checkbox.

Read the full file on GitHub · 78 lines

Changes

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.

  1. today Changed · +68 tokens per session 651f2213d61a
  2. 4d ago First seen · 78 lines · 3,653 tokens per session scan A b37e6ceb9323

Subscribe to this mod's changes

task-pipeline CLAUDE.md is an instructions file published in the GitHub repository SpaiR/task-pipeline (7 stars, last pushed 3d ago), licensed MIT. It adds 3,721 tokens to every session, about $0.0186 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.