Anchor Intake

What it does

When your agent uses it

• A user brings a ticket
• Substantive engineering task brief and wants to begin implementation planning
• Force every fixed rubric element to be accounted
• Cite present/weak claims against the brief

Example prompts

• “/anchor-intake”

Requirements

• Python 3

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

1. 1Capture the brief. Save the user-authored artifact verbatim. Run
2. 2Read the brief in full. Do not fill the rubric from memory or from the
3. 3Account every fixed rubric element. For each of the ten elements, write
4. 4Surface traps and conflicts. Add span-anchored solution-as-goal and
5. 5Ask only dependent questions. Every question must resolve an element or
6. 6Author intent-level steps. Steps are not execution yet. Each step needs an
7. 7Check and repair. Run
8. 8Finalize. Run

What it can do on your machine

Read from SKILL.md and the folder at commit 56bbe71. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/ (Python), which the agent can run.

  Shell commands in SKILL.md call:

  • python3

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Anchor Intake loads about 1.3k tokens when it runs. Until then it costs about 133 tokens; SKILL.md has 480 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~133

When it runs · the whole SKILL.md, loaded when a task matches

~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from lynxlangya/techne at commit 56bbe71, republished under its MIT licence (© lynxlangya). 480 words, ~1,333 tokens.

name

anchor-intake

description

Interrogate a written engineering implementation brief before work starts. Use when a user brings a ticket, PRD, design note, or substantive engineering task brief and wants to begin implementation planning; force every fixed rubric element to be accounted, cite present/weak claims against the brief, surface gaps, solution-as-goal traps, contradictions, and questions, then emit an intent-level plan. Do not use for bare one-line requests, non-engineering artifacts, code diff review, bug fixing, or execution itself.

.techne/plan/<slug>/
  brief.txt          # captured external artifact
  context.json       # computed by init
  intake.json        # authored by the reviewer/planner
  report.json        # computed by check
  plan.json          # computed by finalize
  intakeReport.json  # computed by finalize

{
  "schema": "techne.intake/1",
  "elements": {
    "data-sources": {
      "disposition": "gap",
      "questionIds": ["Q1"]
    }
  },
  "findings": [
    {
      "id": "S1",
      "kind": "solution-as-goal",
      "citation": {"offset": 16, "quote": "Build a Slack bot"}
    }
  ],
  "questions": [
    {
      "id": "Q1",
      "text": "Which data sources should the implementation read?",
      "resolves": ["data-sources"],
      "planDelta": {"adds": ["Bind implementation to the named source"]}
    }
  ],
  "steps": [
    {
      "id": "s1",
      "title": "Implement only the scoped behavior",
      "dependsOnAssumptions": ["goal-and-why"],
      "dependsOnSteps": [],
      "verifiableOutcome": "The scoped behavior can be verified by the acceptance method",
      "terminal": true
    }
  ]
}

© lynxlangya, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 5 other files (scripts) in skills/anchor-intake of lynxlangya/techne.

• SKILL.md
• README-CN.md
• README.md
• eval.md
• reference.md
• scripts/intake_gate.py

Open the folder on GitHubat commit 56bbe71