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 commands/squirrelsoft-dev/dev/specgit clone --depth 1 https://github.com/squirrelsoft-dev/devWhat 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 | $0.00007 | $0.00619 |
| Opus 5 | $0.00003 | $0.00309 |
| Sonnet 5 | $0.00001 | $0.00124 |
| Haiku 4.5 | $0.00001 | $0.00062 |
Grade A, and why
spec 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spec Generator
Generate detailed specification files from a task breakdown.
Input: $ARGUMENTS is the name of a task file (maps to .claude/tasks/$ARGUMENTS.md).
Workflow
- Read the task list file from
.claude/tasks/$ARGUMENTS.md. If it doesn't exist, tell the user and suggest running/breakdown $ARGUMENTSfirst. - Parse all tasks — items matching the pattern
- [ ] **Task title**. - Create the output directory
.claude/specs/$ARGUMENTS/. - Use
TeamCreateto create a team namedspec-$ARGUMENTS. - For each task, spawn an agent using the
Tasktool (subagent_type:general-purpose) withteam_nameset tospec-$ARGUMENTS. Run agents in parallel where possible. - Each agent receives:
- The task title
- The task description (lines following the checkbox until the next task)
- The listed files
- Dependency info (Blocked by / Blocking)
- The full path to write its spec file:
.claude/specs/$ARGUMENTS/{task-title-kebab}.md
- Each agent must:
- Read the files listed in the task to understand existing code
- Search for relevant community skills by running
npx skills find <topic>for the key technologies or patterns in the task. If a useful skill is found, runnpx skills add <owner/repo@skill>to install it and note it in the spec under Implementation Details. - Write a spec file to
.claude/specs/$ARGUMENTS/{task-title-kebab}.mdusing the format below
- After all agents complete, list the generated spec files for the user.
Spec File Format
Each agent should produce a file with this exact structure:
# Spec: <Task Title>
> From: .claude/tasks/{name}.md
## Objective
<What this task accomplishes and why>
## Current State
<Relevant existing code/architecture — agent reads the listed files>
## Requirements
- <Specific, testable requirements>
## Implementation Details
- Files to create/modify with descriptions of changes
- Key functions/types/interfaces to add
- Integration points with existing code
## Dependencies
- Blocked by: <tasks that must complete first>
- Blocking: <tasks that depend on this>
## Risks & Edge Cases
- <Potential issues and mitigations>
## Verification
- <How to confirm this task is done correctly>
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 First seen · 69 lines · 7 tokens per session scan A 15cb8894d691
spec is a command published in the GitHub repository squirrelsoft-dev/dev (14 stars, last pushed 16d ago), licensed MIT. It adds 7 tokens to every session and 619 once invoked, about $0.0000 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 2026-09-01.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.