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/dvnghiem/flowdeck/fd-donegit clone --depth 1 https://github.com/DVNghiem/FlowDeckWhat 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 | $0.00015 | $0.01494 |
| Opus 5 | $0.00008 | $0.00747 |
| Sonnet 5 | $0.00003 | $0.00299 |
| Haiku 4.5 | $0.00002 | $0.00149 |
Grade A, and why
fd-done 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 — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Done
Close the task: summarize what was built against what was asked, then commit and push.
Input: $ARGUMENTS — optional --topic=<slug> to target a topic other than the
active one.
Step 0: Pre-flight
- Check
~/.fd-plan/<slug>/STATE.mdexists. If not:"No planning workspace. Run /fd-task first." - Resolve
<topic>from--topic, else fromtopicin STATE.md. - Read STATE.md via
planning-state action=read. Recordstatus,plan_confirmed,blockers,steps_complete.
Step 1: Verify gate
/fd-done only runs after /fd-verify passed.
If status != "verified":
❌ Cannot close — /fd-verify has not passed for "<topic>".
Run /fd-verify first. If it reported failures, fix them and re-run it.
The agent MUST stop. The agent MUST NOT commit and MUST NOT update state.
The agent MUST also stop if blockers is non-empty, listing each blocker.
Step 2: Summarize built vs required
Read ~/.fd-plan/<slug>/<topic>/task.md and check every requirement and acceptance
criterion against what actually landed.
Collect the changed files:
git diff --name-only HEAD
git log --oneline --no-merges <base>..HEAD
Present:
════════════════════════════════════════════════════
SUMMARY: <topic>
════════════════════════════════════════════════════
Requirements
✅ R-01: <requirement> — <where it landed>
✅ R-02: <requirement> — <where it landed>
⚠️ R-03: <requirement> — not implemented: <reason>
Acceptance Criteria
✅ <criterion>
⚠️ <criterion> — <gap>
Changed files: <N>
<file 1>
<file 2>
════════════════════════════════════════════════════
The agent MUST report gaps honestly. A requirement that was dropped or deferred MUST be stated as such. The agent MUST NOT quietly omit it.
Step 3: Ask for the commit message
Propose one in Conventional Commits form, derived from task.md:
Proposed commit message:
<type>: <description>
<body — what changed and why>
Use this message, edit it, or type SKIP to stop without committing.
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 · 209 lines · 15 tokens per session scan A 4197609ce332
fd-done is a command published in the GitHub repository DVNghiem/FlowDeck (24 stars, last pushed 14d ago), licensed MIT. It adds 15 tokens to every session and 1,494 once invoked, about $0.0001 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
wtfp:contribute
Guide a bounded WTF-P contribution from branch creation through a reviewable change request.
cm
Stage working tree changes and create a Conventional Commit (no push).
mychanges
Summarize my recent git commits in standup format.
checkpoint
Create a checkpoint by staging the changes you have made using git. usage: git [-v | --version] [-h | --help] [-C ] [-c = ] [--exec-path[= ]] [--html-path] [--man-path] [--info-path] [-p | --paginate | -P | --no-pager] [--no-replace-objects] [--bare] [--git-dir= ] [--work-tree= ] [--namespace= ] [--super-prefix= ]…
annotate-plan
Draft a plan and open it in the annotator UI for user review.
refine-title
深度分析 Issue 或 PR 内容，并将其标题重构为 Conventional Commits 格式.