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 skills/raine/consult-llm/phased-implementnpx skills add raine/consult-llm --skill phased-implementgit clone --depth 1 https://github.com/raine/consult-llmWrote 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/skills/raine/consult-llm/phased-implement)<a href="https://agentmods.dev/skills/raine/consult-llm/phased-implement"><img src="https://agentmods.dev/badge/skills/raine/consult-llm/phased-implement.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 | $0.00045 | $0.03735 |
| Opus 5 | $0.00023 | $0.01868 |
| Sonnet 5 | $0.00009 | $0.00747 |
| Haiku 4.5 | $0.00005 | $0.00374 |
Grade A, and why
phased-implement 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.
How it starts
The opening of the file, as written. The whole thing — 459 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Coordinate multi-phase implementation across workmux worktrees. The coordinator owns the master plan, phase dispatch, merge order, ancestry checks, final integration validation, and final summary. The coordinator does not edit source files for feature work.
Do not use Claude Code built-in worktree features. Use workmux for worktree orchestration.
Operating principles
- Coordinator never writes source. All implementation happens inside spawned worktree agents through
/implement. workmux doneis not success. A phase succeeds only when its result sentinel reportsstatus: successand the coordinator verifies it.- Merges are serialized. At most one
/merge --keepruns at a time. - Dependents spawn only after every dependency is merged. No exceptions.
- Drain completed handles before dispatching dependents.
workmux wait --anycan return after one transition while sibling phases also finished. - Use bounded waits. Avoid tight polling and inspect
workmux statuson timeouts. - Preserve failed and blocked worktrees for inspection. Remove worktrees only after merge and ancestry verification succeed.
- Treat merge conflicts as merge failures for that phase, not as reasons to destroy the worktree.
- Read the master plan YAML semantically. Do not write shell parsers for YAML.
- Keep all coordinator artifacts in one shared
history/<date>-phased-<slug>/run directory so worktrees can read prompts and write sentinels.
Argument handling
Arguments are $ARGUMENTS.
Parse these flags before starting:
--plan <path>: load an existing master plan instead of generating one.--integration-branch <branch>: branch that receives phase merges. Default: current branch.--preset light|standard|design|strict: default preset for generated phases.--validation <command>: final integration validation command.--reviewer <selector>: consult-llm selector for optional integration review. Repeatable.--reviewers <selector,selector>: comma-separated reviewer selectors.--integration-review none|auto|full: final integration review policy. Default:auto.
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.
- today Changed · -26 lines 505dd03881d8
- 5d ago First seen · 485 lines · 45 tokens per session scan A 605b78102aba
phased-implement is a skill published in the GitHub repository raine/consult-llm (132 stars, last pushed today), licensed MIT. It adds 45 tokens to every session and 3,735 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 2026-08-30.
Other skills, from other repositories
agile-product-owner
../../../product-team/agile-product-owner/skills/agile-product-owner/SKILL.md.
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
story-readiness
Validate that a story file is implementation-ready. Checks for embedded GDD requirements, ADR references, engine notes, clear acceptance criteria, and no open design questions. Produces READY / NEEDS WORK / BLOCKED verdict with specific gaps. Use when user says 'is this story ready', 'can I start on this story', 'is…
flow-next-tracker-sync
Project a flow-next spec to a tracker issue (Linear, GitHub, GitLab, Jira) and reconcile two-way. Use when asked to sync to a tracker. NOT plan-sync.
loop
Full execution protocol for MODE: LOOP — the compound-engineering loop: brainstorm → plan → build → review → improve, iterating under defense-in-depth stop conditions with generator/critic separation, durable resumable state, and mandatory compounding learning capture. Loaded on demand by the architect when the loop…
subagent-delegation
Canonical protocol for delegating GSD work to native Antigravity subagents — when to delegate, how to invoke, workspace isolation modes, and the inline fallback for older IDE versions.