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 skills add irfad7/claude-power-skills --skill thinkbackgit clone --depth 1 https://github.com/irfad7/claude-power-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/skills/irfad7/claude-power-skills/thinkback)<a href="https://agentmods.dev/skills/irfad7/claude-power-skills/thinkback"><img src="https://agentmods.dev/badge/skills/irfad7/claude-power-skills/thinkback/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/irfad7/claude-power-skills/thinkback"><img src="https://agentmods.dev/badge/skills/irfad7/claude-power-skills/thinkback.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00099 | $0.01803 |
| Opus 5 | $0.00049 | $0.00901 |
| Sonnet 5 | $0.00020 | $0.00361 |
| Haiku 4.5 | $0.00010 | $0.00180 |
Grade A, and why
thinkback 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 10d 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 — 241 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Thinkback — Decision Archaeology & Reasoning Replay
You are a decision archaeologist. You dig up past decisions, reconstruct the reasoning that led to them, evaluate whether those reasons still hold, and extract lessons for future decision-making.
Why Thinkback Exists
Every codebase is a fossil record of decisions. Most of those decisions made sense when they were made — but the context that justified them is lost. Developers inherit code and either: (a) blindly preserve decisions they don't understand, or (b) blindly undo decisions that were actually good. Thinkback prevents both.
Three Modes
Mode 1: Decision Archaeology (/thinkback dig [topic])
Reconstruct the full reasoning chain behind a past decision.
The Dig Process:
Step 1: Evidence Collection Gather every trace of the decision:
Evidence Sources:
1. Git blame — who wrote this code, when, what commit message?
2. Git log — what changed around the same time?
3. PR history — was there a PR? What was discussed?
4. Comments in code — any TODO, HACK, FIXME, NOTE annotations?
5. Decision files — LINEAGE.md, ADRs, design docs?
6. Memory files — MEMORY.md, SCARS.md references?
7. The code itself — what does the implementation reveal about intent?
8. Dependencies added — what was introduced around the same time?
9. Config changes — did environment or settings change?
10. Test files — what was tested (and what wasn't)?
Step 2: Context Reconstruction Build a picture of the world when the decision was made:
DECISION CONTEXT:
═══════════════════
When: [date from git blame/log]
Who: [author]
What existed then:
- [tech stack at the time]
- [scale/usage at the time]
- [team size/skills at the time]
- [constraints that may no longer exist]
What didn't exist yet:
- [features built after this decision]
- [tools/libraries that now exist]
- [requirements that came later]
Step 3: Reasoning Reconstruction Based on the evidence, reconstruct the likely reasoning:
RECONSTRUCTED REASONING:
═══════════════════════
The decision was: [what was decided]
Likely because:
1. [First reason — evidence: commit message / code pattern / timing]
2. [Second reason — evidence: dependency added / test written]
3. [Third reason — evidence: config change / comment in code]
Alternatives they likely considered:
- [Alternative A] — probably rejected because [reason]
- [Alternative B] — probably rejected because [reason]
Confidence in this reconstruction: [HIGH / MEDIUM / LOW]
- [Why this confidence level — what evidence is strong/weak]
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.
- 10d ago First seen · 241 lines · 99 tokens per session scan A b670cf0f0fdd
thinkback is a skill published in the GitHub repository irfad7/claude-power-skills (4 stars, last pushed 1mo ago), licensed MIT. It adds 99 tokens to every session and 1,803 once invoked, about $0.0005 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 skills, from other repositories
report
Writes the session final report to a file, then prints only the path and a one-line summary. Fires when the prompt contains "Report per memstack:report", and also when the prompt begins with a standing trigger configured through MEMSTACKREPORTONTASKPROMPTS or MEMSTACKREPORTTRIGGERS. Dormant otherwise.
compress
Use when the user says 'tokenstack', 'compression', 'token savings', 'proxy status', or asks about context window usage.
token-optimization
Use when the user says 'token optimization', 'save tokens', 'context window', 'reduce tokens', 'token stack', or 'TokenStack', or asks about extending context window capacity. Covers TokenStack, the built-in compression proxy that shrinks Claude Code tool output before it reaches the Anthropic API. Do NOT use for…
state
Use when the user says 'update state', 'project state', 'where was I', or at session start to load current context.
grimoire
Use when the user says 'update context', 'update claude', 'save library', or after significant project changes.
project-butler
A project-memory workflow that keeps a code project’s rules, current structure, documents, logs, tasks, and handoff notes organized.