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/kam-l/claude-coachWrote 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/kam-l/claude-coach/reflect)<a href="https://agentmods.dev/commands/kam-l/claude-coach/reflect"><img src="https://agentmods.dev/badge/commands/kam-l/claude-coach/reflect.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 | $0.00041 | $0.00889 |
| Opus 5 | $0.00020 | $0.00445 |
| Sonnet 5 | $0.00008 | $0.00178 |
| Haiku 4.5 | $0.00004 | $0.00089 |
Grade A, and why
reflect 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.
How it starts
The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review pending session reflections — proposed memories, tips, CLAUDE.md patches, and skill patches extracted from past sessions.
Steps
-
Read pending reflections from the project JSONL:
- Derive project slug from the current working directory: strip trailing slashes, replace
:,\,/with- - Resolve
$HOMEto the absolute home path. Read{home}/.claude/projects/{slug}/pending-reflections.jsonl(one JSON per line) - If file doesn't exist or is empty, tell the user and exit
- Derive project slug from the current working directory: strip trailing slashes, replace
-
For each pending reflection, parse the JSON and print as regular text output (not inside AskUserQuestion):
- Session timestamp and working directory
- Extracted signals as a numbered list with tier label (correction / approval / observation)
- Each proposed item as a labeled line:
[type] name — description
-
If
$ARGUMENTSis "accept-all", apply all reflections without prompting (batch mode). -
Otherwise, use one
AskUserQuestioncall batching up to 4 reflections:- One multi-select question per reflection (max 4 questions per call)
- Question: "Session {date} — select items to accept:", header: date string (e.g.
"Mar 31") - Each proposed item is an option: label=
[type] name, description=content summary - User selects which items to accept; unselected items are skipped
- No separate cherry-pick flow — the multi-select IS the cherry-pick
- If >4 reflections pending, process in sequential batches of 4
-
For accepted memories:
- Derive project slug from the reflection's
cwdfield: strip trailing slashes, replace:,\,/with-(e.g.,C:\Projects\foo→C--Projects-foo) - Write memory file to
{home}/.claude/projects/{project-slug}/memory/{name}.md(resolve$HOMEto absolute path — Write tool doesn't expand tilde) - Use frontmatter format:
name,description,typefields - Update
{home}/.claude/projects/{project-slug}/memory/MEMORY.mdindex with a link to the new file - Check for duplicates against existing MEMORY.md entries before writing
- Derive project slug from the reflection's
-
For accepted tips:
- Display the tip and suggest the user run
/setup refreshto include it
- Display the tip and suggest the user run
-
For accepted CLAUDE.md patches:
- Read the project's CLAUDE.md (from the reflection's
cwd) - Find the target
## Sectionheading - Append the patch content under that section (before the next
##heading) - If section doesn't exist, create it at the appropriate position
- Show the diff to the user before writing
- Read the project's CLAUDE.md (from the reflection's
-
For accepted skill patches:
- Find the skill at
skills/{skill_name}/SKILL.md(check both project and{home}/.claude/skills/) - Read the file, find the target section
- Append the patch content under that section
- If the skill or section doesn't exist, show the content and ask where to put it
- Show the diff to the user before writing
- Find the skill at
-
After processing, truncate
pending-reflections.jsonl(clear the file).
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 · 66 lines · 41 tokens per session scan A 7950c827a49a
reflect is a command published in the GitHub repository kam-l/claude-coach (11 stars, last pushed 5mo ago), licensed MIT. It adds 41 tokens to every session and 889 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
minutes-ideas
Surface recent voice memos and ideas captured from any device. Use when the user asks "what ideas did I have?", "what were my recent memos?", "what did I record while walking?", or wants to recall a captured thought.
learn
Force claude-smart to extract learnings from this session now.
memory-store
Store an insight, decision, or pattern to memory.
cc-memory
Configure persistent memory that survives across sessions using a layered approach: split rule files for always-loaded context, auto-memory for organic learning, and optional MCP-backed long-term memory for large codebases.
analyze-context
USE WHEN you want to analyze project context before starting work on a task. Calls context + recall, then synthesizes goals, decisions, gotchas, and relevant memories into a pre-task brief.
music-suno-prompt
Grounded Suno prompt synthesis from local knowledge corpus + persona canon + label canon. No vibes-prompting.