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 softspark/ai-toolkit --skill predictgit clone --depth 1 https://github.com/softspark/ai-toolkitWrote 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/softspark/ai-toolkit/predict)<a href="https://agentmods.dev/skills/softspark/ai-toolkit/predict"><img src="https://agentmods.dev/badge/skills/softspark/ai-toolkit/predict/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/softspark/ai-toolkit/predict"><img src="https://agentmods.dev/badge/skills/softspark/ai-toolkit/predict.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00039 | $0.01176 |
| Opus 5 | $0.00019 | $0.00588 |
| Sonnet 5 | $0.00008 | $0.00235 |
| Haiku 4.5 | $0.00004 | $0.00118 |
Grade A, and why
predict 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 5d 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Predict Command
$ARGUMENTS
Triggers the Predictive Analyst to assess the impact and regression risk of proposed changes.
Usage
/predict [path_or_diff]
# /predict src/auth : analyze all files under src/auth
# /predict --diff : analyze uncommitted changes (git diff)
# /predict src/api/routes.ts : analyze a single file
Protocol
1. Scope: Identify Target Files
- If path provided: collect all files under that path
- If
--diff: rungit diff --name-onlyto get changed files - List each file with its last-modified date and line count
2. Trace: Build Dependency Graph
For each target file, find dependents:
# Find files that import/require the target
grep -rl "import.*from.*[target]" --include="*.ts" --include="*.py" --include="*.js" .
grep -rl "require.*[target]" --include="*.js" --include="*.ts" .
Build a graph: changed file, direct dependents, transitive dependents (1 level)
3. Assess: Calculate Risk Score
Score each changed file on a 1 to 5 scale:
| Factor | Weight | Scoring |
|---|---|---|
| Dependent count | 30% | 0 deps = 1, 1 to 3 = 2, 4 to 10 = 3, 11 to 20 = 4, 21+ = 5 |
| Test coverage | 30% | Has dedicated test = 1, partial = 3, none = 5 |
| Change surface | 20% | < 10 lines = 1, 10 to 50 = 2, 50 to 200 = 3, 200+ = 5 |
| Shared/core file | 20% | Leaf = 1, mid-layer = 3, core/shared = 5 |
Overall risk = weighted average rounded to nearest integer.
4. Report: Generate Impact Prediction
Output a markdown report:
## Impact Prediction: [scope]
| File | Risk | Dependents | Test Coverage | Notes |
|------|------|------------|---------------|-------|
| src/auth/login.ts | 4/5 | 12 files | partial | Core auth flow |
### High-Risk Changes (score >= 4)
- [file]: [why it's high risk and what to watch]
### Recommended Actions
- [ ] Add tests for [untested file]
- [ ] Review [high-dependent file] with extra scrutiny
- [ ] Run integration tests covering [affected area]
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.
- 5d ago First seen · 104 lines · 39 tokens per session scan A 2c13a18d34b8
predict is a skill published in the GitHub repository softspark/ai-toolkit (170 stars, last pushed today), licensed Apache-2.0. It adds 39 tokens to every session and 1,176 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 2026-09-03.
Other skills, from other repositories
cline-fix-volatile-msg
Ladder-aware Cline Anthropic caching — verify the rolling read/write ladder on the wire, then add the tools breakpoint and tune TTL. Updated for the 2026-08 AI-SDK monorepo.
continue-fix-volatile-msg
Ladder-aware Continue Anthropic caching — verify the rolling ladder on the wire, then enable it by default and add TTL coverage.
opencode-detect-openai-compat
OpenCode's caching detection misses OpenAI-compatible proxies routing to Anthropic/Bedrock. Broaden the predicate.
cline-pin-timestamp
Cline's system prompt includes a timestamp that may be recomputed per request, invalidating the system-prompt cache.
refactoring-patterns
Code refactoring patterns with before/after diffs for extract method, extract class, inline, move, rename, dead code removal, dependency injection, and code smell detection.
review-loop
Run the adversarial verification loop — implement, then hand the change to a fresh checker that did not write it, fix what it finds, and re-dispatch until APPROVE. Use before claiming any behavioural change is done, and on requests like "review loop", "adversarial review", "independent review", "get this verified"…