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.
git clone --depth 1 https://github.com/CrashBytes/claude-role-skillsWrote 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/commands/crashbytes/claude-role-skills/retro)<a href="https://agentmods.dev/commands/crashbytes/claude-role-skills/retro"><img src="https://agentmods.dev/badge/commands/crashbytes/claude-role-skills/retro.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 5.1 | $0.00006 | $0.00529 |
| Opus 5 | $0.00003 | $0.00264 |
| Sonnet 5 | $0.00001 | $0.00106 |
| Haiku 4.5 | $0.00001 | $0.00053 |
Grade A, and why
retro 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 7d 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.
What it actually says
Sprint Retrospective
Facilitate a structured sprint retrospective.
Arguments
Requested format: $ARGUMENTS
Workflow
-
Select format
- If a format is specified, use it
- If not specified, default to Start/Stop/Continue (most universal)
- Available formats: start-stop-continue, 4ls, sailboat, timeline, fishbone
-
Set the stage (2 min)
- State the Prime Directive: "Regardless of what we discover, we understand and truly believe that everyone did the best job they could, given what they knew at the time, their skills and abilities, the resources available, and the situation at hand."
- If a project management MCP server is connected, pull the sprint metrics (velocity, burndown, completed vs committed)
-
Run the format
Start/Stop/Continue:
- Ask: What should we START doing? What should we STOP doing? What should we CONTINUE doing?
- Generate 3-5 prompting questions for each category based on common team patterns
4Ls (Liked, Learned, Lacked, Longed For):
- Ask for items in each L category
- Group related items across categories
Sailboat:
- Wind (what propels us forward), Anchor (what holds us back), Rocks (risks ahead), Island (our goal)
- Map items to actionable improvements
Timeline:
- Walk through the sprint week by week
- Mark highs, lows, and key events
- Identify patterns across the timeline
Fishbone / 5 Whys:
- Identify the main problem from the sprint
- Apply 5 Whys to find root cause
- Categorize causes: People, Process, Tools, Environment
-
Generate action items
- Each action item must have: owner, due date, and success criteria
- Limit to 3 action items (more won't get done)
- Reference previous retro actions if available
-
Output the retro summary
- Format chosen and why
- Key themes identified
- Top 3 action items with owners
- Team health pulse (improving / stable / declining)
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.
- 7d ago First seen · 60 lines · 6 tokens per session scan A 29560e95bf04
retro is a command published in the GitHub repository CrashBytes/claude-role-skills (6 stars, last pushed 6mo ago), licensed Apache-2.0. It adds 6 tokens to every session and 529 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 2026-08-31.
Other commands, from other repositories
contribute-to
Join an EXISTING, in-progress work item (issue or PR) as a contributor — walk the contribution loop toward a human-stated goal and success criteria, without the full spec chain.
create-ticket
Create a work-item ticket (GitHub issue / Jira) from an existing requirements.md, then promote its draft spec folder to docs/specs/ /.
finish-tasks
Finalize a work item after all tasks are done — cleanup (currently closing the ticket) and mark the work complete.
work-status
Report the current status of a work item by reading its spec files (requirements/design/tasks) and execution log. Read-only.
ledger-check
Audit all ledger files for drift (report; fix only on request).
triage
Promote ideas from Ideas.md into Epics or User Stories in the Epic backlog (dedup, typed ID, voice).