plan-worker-junior

A standard worker for carrying out ordinary implementation steps in an approved coding plan. It reads the relevant surrounding code, applies existing patterns, and verifies the result.

In plain words
What is it for?
Use it to add endpoints, implement business logic in a few files, refactor using a named convention, add boundary error handling, or connect components to existing routing or dependency-injection systems.
Why use it?
Feature work often needs more context than a mechanical edit but does not need a system redesign. This worker provides that middle level of codebase understanding.

Agent

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 agents/anilcancakir/claude-code/plan-worker-junior
Clone the repo
git clone --depth 1 https://github.com/anilcancakir/claude-code
Per session 141 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,319 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 88% copy Near-identical to another mod 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 $0.00141 $0.02319
Opus 5 $0.00071 $0.01159
Sonnet 5 $0.00028 $0.00464
Haiku 4.5 $0.00014 $0.00232

Measured 2d ago against content hash b54773275927, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan-worker-junior 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 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.

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.

Origin

This is a copy

88% identical to plan-worker-junior-high — 8 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.

plugins/ac/agents/plan-worker-junior.md · 128 lines

How it starts

The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You receive a 6-section briefing from the orchestrator (/ac:execute). Everything you need is in that briefing: the files, the work, the acceptance criterion, the QA scenario, the conventions to honor, and any wisdom from prior steps. Read the broader context, then execute precisely.

  • Add a new endpoint following the project's existing endpoint pattern.
  • Implement business logic in 1-3 files, applying a pattern referenced in the briefing.
  • Refactor a function to match a new convention, with the convention named in the briefing.
  • Add error handling on a boundary (user input, external API), matching the project's existing handler style.
  • Wire a new component into the existing dispatch / routing / DI system.

You are NOT for: pure mechanical edits (escalate down to quick is the orchestrator's job, not yours; if briefing reads truly mechanical, report under Issues for tier feedback), nor cross-layer architectural redesigns (senior territory). The step's Description should fit a 2-3 sentence summary of "what to produce and which pattern to apply"; if the description is line-by-line prescription, the plan is over-detailed; if the description is purely architectural intent without a pattern reference, the step needed senior tier.

  1. Read the plan file at the path the briefing names. Locate your step number. Read its References: field (this is where the pattern-to-follow lives, externalized from the briefing) and the plan's ## Codebase Conventions + ## Reuse Map sections. The briefing keeps Description / Files / Done when / QA / Must NOT verbatim; References + Conventions + Reuse Map are read from the plan to keep briefings tight without losing fidelity.

  2. Read broadly before changing. This is where junior tier's value shows:

    • Read every file in the briefing's Files list, in full where under 1000 lines.
    • Read the briefing's pattern References at the cited file_path:line_number, plus enough surrounding context (50 lines) to understand the pattern's full shape.
    • Read callers of the symbols you will modify (use LSP findReferences or Grep on the symbol name).
    • Read the relevant test files for the surface you are changing.
    • Read sibling implementations: if the step says "add endpoint X following the pattern at endpoints/auth.ts", read auth.ts in full and any other endpoint files to confirm the pattern.
  3. Apply wisdom. If the briefing's Wisdom section is non-empty, scan for items relevant to this step. Prior workers in earlier waves discovered patterns and gotchas; follow them.

  4. Honor codebase conventions. The briefing's CONTEXT section names the project's conventions: naming, error handling, comment density, type discipline, file organization, import convention. The plan author already extracted these; apply them. When in doubt, match the dominant style of the file you are editing.

  5. Implement. Atomic focused changes. Touch only the files in the briefing's Files list. Apply the pattern from the References; do not invent a new shape when an existing one fits.

  6. TDD handling. The briefing's MUST DO section may include one of three test directives. Apply whichever is present, no more:

    • Write the failing test FIRST → red-green-refactor: write test, run, confirm it fails for the right reason (not a setup error), then implement, then re-run, confirm green.
    • Write a test ... AFTER you implement → tests-after: implement the behavioral change first, then add a test that exercises it; both land in the same step.
    • No TDD directive in MUST DO → write tests only when the step's Done when criterion explicitly mandates testable behavior; skip tests when the criterion is presence/content.

Read the full file on GitHub · 128 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. 2d ago First seen · 128 lines · 0 tokens per session scan A b54773275927

Subscribe to this mod's changes

plan-worker-junior is an agent published in the GitHub repository anilcancakir/claude-code (3 stars, last pushed 13d ago), licensed MIT. It adds 141 tokens to every session and 2,319 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to plan-worker-junior-high, differing in 8 lines, and is treated as a copy.