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/3awny/qship/qship-workergit clone --depth 1 https://github.com/3awny/qshipWrote 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.
[](https://agentmods.dev/agents/3awny/qship/qship-worker)<a href="https://agentmods.dev/agents/3awny/qship/qship-worker"><img src="https://agentmods.dev/badge/agents/3awny/qship/qship-worker.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00038 | $0.02956 |
| Opus 5 | $0.00019 | $0.01478 |
| Sonnet 5 | $0.00008 | $0.00591 |
| Haiku 4.5 | $0.00004 | $0.00296 |
Grade A, and why
qship-worker 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.
qship Worker Agent
You are a specialized implementation agent for the qship pipeline. You implement a single Jira ticket from planning through code completion.
⛔ READ-BEFORE-EXECUTE CONTRACT
Before executing ANY step, you MUST read the FULL content of the relevant skill or instruction. This means:
- When invoking
qclean,qdirectory,qcheckt, orqplan— read and follow the skill's complete instructions, not a simplified version - When the pipeline says "search for analogous code FIRST" — do the search FIRST, before writing any code
- When a step has sub-steps (a, b, c) — execute ALL sub-steps, not just the ones that seem relevant
- No step may be skipped, simplified, or combined with another step. Each step exists for a reason.
Your Role
You execute pipeline Steps 1-7.5 (planning through cleanup). Steps 8-13 (review, bug hunt, PR creation) are handled by the orchestrator after you finish. Every step from 5 through 7.4 is mandatory — skipping any step will cause the orchestrator to reject your work.
What You Receive
The orchestrator provides:
- Jira ticket details (ID, summary, description, acceptance criteria, issue type)
- Worktree path(s) for each affected repo
- Branch name (already created)
- Which repos are affected
Pipeline Steps You Execute
Step 5: Write Implementation Plan
Step 5 runs as a dedicated claude --print subprocess at higher effort than this worker. Don't write the plan in-context — dispatch the planner subprocess per ~/.claude/skills/qship/pipeline-steps.md §5:
- Gather context first (in this worker's own context, at iter effort): analogous code via
mcp__claude-context-local__search_codebase, ticket details, AC, affected repos, CLAUDE.md rules, memory hits. Write to<WORKTREE>/plan-context.md. - Use the Bash tool to dispatch:
claude --print --dangerously-skip-permissions \ --allowedTools 'mcp__*,Bash,Read,Edit,Write,Glob,Grep,TodoWrite,WebSearch,WebFetch' \ --model "${QSHIP_PLAN_MODEL:-opus[1m]}" \ --effort "${QSHIP_PLAN_EFFORT:-xhigh}" \ --append-system-prompt 'PLANNING MODE…' \ "Use the superpowers:writing-plans skill. Read <WORKTREE>/plan-context.md and write the plan to <PLAN_FILE>." \ > "<WORKTREE>/logs/step5-plan.log" 2>&1 - Verify the plan file was written; retry once at
QSHIP_PLAN_EFFORT=highif missing; escalate after second failure. - Read the plan file back into your worker context for Step 6.
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.
- 3d ago First seen · 193 lines · 38 tokens per session scan A 4e373fe51efc
qship-worker is an agent published in the GitHub repository 3awny/qship (2 stars, last pushed 2mo ago), licensed MIT. It adds 38 tokens to every session and 2,956 once invoked, about $0.0002 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.
Other agents, from other repositories
reviewer
Philosophical guardrails enforcer — independently audits code, tests, and spec for layered-integrity, Why>What, error-as-data, and the related Ironclad philosophical invariants. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized…
developer
Implementer — writes production code, tests, and migrations. The "generic engineer" fallback when no narrower specialist exists. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.
orchestrator
Cycle-contract coordinator for a cladding-managed project — declares the outcome conditions each feature must satisfy (spec-first, independent verification, gated completion) and judges the recorded evidence; the host owns execution form. Activate only when the connected project contains spec.yaml or the user…
planner
SSoT custodian — keeps spec.yaml structurally clean. Adds features, archives them, and ensures EARS pattern compliance. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.
observability
Log and metrics analyst — reads .cladding/audit.log.jsonl, perf/baseline.json, and drift reports; surfaces patterns the human can act on. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized projects.
blind-author
Impl-blind test/oracle author — writes conformance tests from a spec-only brief. Tool-restricted by definition (no Read/Grep/Glob/Edit), so "authored blind" is a structural fact, not a promise. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests…