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/joanseg/specmanager/specmanager-designgit clone --depth 1 https://github.com/joanseg/specmanagerWhat 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 | $0.00044 | $0.00726 |
| Opus 5 | $0.00022 | $0.00363 |
| Sonnet 5 | $0.00009 | $0.00145 |
| Haiku 4.5 | $0.00004 | $0.00073 |
Grade A, and why
specmanager-design 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.
How it starts
The opening of the file, as written. The whole thing — 30 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design the screens for the feature: $ARGUMENTS.
Steps
-
Resolve the feature. Call
list_features. Find the feature whoseidorslugmatches the first whitespace-separated token of the argument. If none match, ask the user to clarify and stop. (Anything after the first token is treated as extra context to forward to the subagent.) -
Check the gate. Call
check_gate({ featureId, stage: "design" }). The design gate is open once the PRD is approved (Architecture is NOT required — design can run in parallel with Architecture). If closed, reportreasonand stop. -
Confirm no draft already exists.
list_documents({ featureId, stage: "design" }). If a doc exists, point the user at the panel to edit, or delete the file before regenerating. Don't duplicate. -
Look up PRD and Architecture ids to pass into the subagent prompt.
list_documents({ featureId, stage: "prd" })andlist_documents({ featureId, stage: "architecture" }). Architecture may be absent or in draft — passnullif so. -
Collect attachments. If the user pasted screenshot paths in the conversation before invoking the command, list them — the designer uses them as visual reference and may inline them as data URIs. If no attachments, the designer works from the PRD + DESIGN.md alone. Optional design reference (R5/AC7): when
docs/DESIGN.mdholds only placeholder/# TODOtokens, the designer may invite an optional design reference (a screenshot/example through this same attachment path) to ground a synthesized token system — always optional, never required; the designer proceeds and synthesizes if none is given. -
Invoke the subagent. Use the
Tasktool withsubagent_type: "designer"and a prompt that includes:- The feature id, title, and slug.
- The PRD id + version.
- The Architecture id + version (if any).
- The screenshot paths the user attached (if any).
- Any extra context the user gave after the feature id.
The designer reads the upstream docs and
./docs/DESIGN.md, designs the actual screens as one self-contained HTML file (high-fi mockups stacked with explanatory notes between them), and callscreate_design_briefitself. -
Sync CLAUDE.md. After the subagent returns, call
sync_claude_md. -
Report. Document id + file path (
design/mockups.html). Suggest opening it in the board — the doc panel renders the stacked mockups in a sandboxed iframe preview.
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.
- 2d ago First seen · 30 lines · 44 tokens per session scan A ec3aad513f8c
specmanager-design is a command published in the GitHub repository joanseg/specmanager (19 stars, last pushed 18d ago), licensed MIT. It adds 44 tokens to every session and 726 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 commands, from other repositories
example-only
This command documents how to dispatch. Example only.
develop-configure
The develop orchestrator body: phases, dispatch discipline, quality gates, ceremony lock, ledger, and handoff. Loaded by the develop skill once the operator has chosen a ceremony path.
add-skill
Add a new Skill to the current plugin with proper structure.
my-cmd
Use the agent tool with.
bad-cmd
Use the agent tool with.
_subagent-discovery
Whenever a command needs to pick a subagent for a task (planning, task generation, or implementation execution). Do not hardcode agent names in command templates. Do not assume a specific agent exists.