Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/idoforgod/Dissertation-Simulator-AgenticWorkflownpx agentmods add commands/idoforgod/dissertation-simulator-agenticworkflow/self-improveWrote 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/idoforgod/dissertation-simulator-agenticworkflow/self-improve)<a href="https://agentmods.dev/commands/idoforgod/dissertation-simulator-agenticworkflow/self-improve"><img src="https://agentmods.dev/badge/commands/idoforgod/dissertation-simulator-agenticworkflow/self-improve/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/commands/idoforgod/dissertation-simulator-agenticworkflow/self-improve"><img src="https://agentmods.dev/badge/commands/idoforgod/dissertation-simulator-agenticworkflow/self-improve.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 | $0.00019 | $0.01321 |
| Opus 5.5 | $0.00008 | $0.00528 |
| Sonnet 5.5 | $0.00004 | $0.00264 |
| Haiku 4.5 | $0.00002 | $0.00132 |
Grade A, and why
self-improve 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 1mo 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 — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/self-improve — Knowledge-Based Self-Improvement (KBSI)
Analyzes recent workflow execution for errors and improvements, extracts generalizable insights, and permanently stores them in AGENTS.md §11.
Uses a P1-sandwich architecture to prevent hallucinations:
P1 Validation (SI-1~SI-6) → LLM Analysis → P1 Application (marker-based append)
SOT: self-improvement-logs/state.json
Insights: AGENTS.md §11.4 (within SELF-IMPROVEMENT-START/END markers)
Track 2 queued changes: self-improvement-logs/queued-changes/
Orchestration (main context executes — do NOT delegate to sub-agent)
Announce to user: "Starting KBSI self-improvement analysis."
Step 1: Gather Context (P1)
Check KBSI system status:
python3 .claude/hooks/scripts/self_improve_manager.py \
--status --si-dir self-improvement-logs
Report current state to user (applied/pending/rejected counts).
Step 2: Analyze Knowledge Index
Read the knowledge-index.jsonl to identify recurring error patterns:
.claude/context-snapshots/knowledge-index.jsonl
Look for:
- Error patterns that appear ≥ 2 times across sessions (
error_patternsfield) - Rejected hypotheses that recur (
rejected_hypothesesfield) - Success patterns that could be generalized (
success_patternsfield)
Step 3: Generate Insights (LLM)
For each candidate pattern, formulate an insight with:
- Title: Short descriptive name
- Condition: When does this rule apply? (specific, testable)
- Rule: What should be done? (actionable, unambiguous)
- Rationale: Why? (with evidence from KI data)
- Type: SAFE (additive) or STRUCTURAL (changes core behavior)
- Error Type: Associated error category (if Track 1)
Quality criteria — ALL 4 must pass:
- Recurrence: Pattern appeared ≥ 2 times
- Generalizability: Applies beyond the specific instance
- Actionability: Produces a concrete, testable rule
- Non-redundancy: Not already covered by existing §11 insights
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.
- 1mo ago First seen · 170 lines · 19 tokens per session scan A 7fdfb29446a5
self-improve is a command published in the GitHub repository idoforgod/Dissertation-Simulator-AgenticWorkflow (108 stars, last pushed 4mo ago), licensed MIT. It adds 19 tokens to every session and 1,321 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-03.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
speckit.bug.fix
Apply the remediation from a bug assessment and record what was changed.