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/xiaolai/nlpm/trendgit clone --depth 1 https://github.com/xiaolai/nlpmWhat 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 | $0.00014 | $0.00641 |
| Opus 5 | $0.00007 | $0.00320 |
| Sonnet 5 | $0.00003 | $0.00128 |
| Haiku 4.5 | $0.00001 | $0.00064 |
Grade A, and why
trend 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 2d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
User Input
$ARGUMENTS
Workflow
Step 1: Load History
Read .claude/nlpm-history.json from the project root.
- If it doesn't exist: this is the first run. Score everything in Step 2 and the snapshot saved in Step 4 becomes the baseline; Step 3's comparison is skipped (no prior data).
- If it exists but parses as malformed JSON: warn one line, treat as empty, continue.
Step 2: Score Current State
Dispatch the nlpm:scorer and nlpm:vague-scanner agents in parallel to score all artifacts (or artifacts at the given path).
Step 3: Compare Against History
Filter the loaded snapshots to only those whose scope matches the current scope — otherwise a path-bound trend would be compared against full-repo baselines and produce nonsense deltas. The scope is derived from the current invocation's arguments using the same mapping as commands/shared/append-history.md.
For each artifact in the current score:
- Find its most recent entry in the filtered history
- Compute delta: current_score − historical_score
- Flag: improved (delta > 0), degraded (delta < 0), unchanged (delta == 0), new (no history)
If the filtered history is empty (first run for this scope), skip the delta computation and label every artifact new.
Step 4: Save Snapshot
Persist this run by following commands/shared/append-history.md with the scope determined in Step 3, the per-file scores from Step 2, and the file count. The partial handles file creation, deduplication, and atomic write.
Step 5: Report
NLPM Trend Report
Snapshot: 2026-03-28 (3rd snapshot, 2 previous)
File Score Previous Delta
--------------------------------------------------------------
agents/scorer.md 95 92 +3 improved
agents/scanner.md 90 90 0 unchanged
commands/score.md 95 88 +7 improved
skills/nlpm/scoring/SKILL.md 85 85 0 unchanged
.claude/rules/testing.md 78 82 -4 degraded
commands/fix.md (NEW) 88 -- new
Overall: 88/100 (was 87, +1)
Degraded (needs attention):
.claude/rules/testing.md 82 → 78 (-4)
Trend: 3 snapshots — 82 → 87 → 88 (improving)
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.
- 2d ago First seen · 71 lines · 14 tokens per session scan A 59df9900ba76
trend is a command published in the GitHub repository xiaolai/nlpm (133 stars, last pushed 2d ago), licensed ISC. It adds 14 tokens to every session and 641 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
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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.
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.