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 Parcha-ai/parcha-skills --skill parablegit clone --depth 1 https://github.com/Parcha-ai/parcha-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/parcha-ai/parcha-skills/parable)<a href="https://agentmods.dev/skills/parcha-ai/parcha-skills/parable"><img src="https://agentmods.dev/badge/skills/parcha-ai/parcha-skills/parable/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/parcha-ai/parcha-skills/parable"><img src="https://agentmods.dev/badge/skills/parcha-ai/parcha-skills/parable.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 360 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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 | $0.00115 | $0.05801 |
| Opus 5.5 | $0.00046 | $0.02320 |
| Sonnet 5.5 | $0.00023 | $0.01160 |
| Haiku 4.5 | $0.00012 | $0.00580 |
Grade A, and why
parable scanned grade A with 1 finding 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 2mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
it with `curl`, `npx`, or a repository clone: the skill carries the pinned runtime it needs. How it starts
The opening of the file, as written. The whole thing — 363 lines — stays where its author put it; the contents beside it link to each section on GitHub.
parable — divide and conquer
First-time install and subscription onboarding
When the user asks to install, set up, or onboard Parable, resolve parable.sh beside this
SKILL.md. Plugin caches and standalone skill installs live in different roots. Do not replace
it with curl, npx, or a repository clone: the skill carries the pinned runtime it needs.
Before running the script, state that Claude is the baseline subscription because the harness is Claude Code, then collect the optional subscriptions with the harness's native structured-question UI:
- Ask whether to connect a ChatGPT subscription for Sol, Terra, and Luna. Explain that Sol becomes
an automatic fallback only when ChatGPT is connected; without it,
autoremains on Fable. - Ask whether to connect an xAI subscription for Grok 4.5. Recommend yes when the user has one.
- Ask whether to connect a Kimi Code subscription for Kimi. Recommend yes when the user has one.
In Claude Code use AskUserQuestion; in Codex use request_user_input when the active mode
permits it (normally Plan mode); in another harness use its equivalent structured-choice tool.
Ask all three questions in one tool call when possible. Use two-choice yes/no answers and explain
that the credentials stay user-owned in CLIProxyAPI. If the structured-question tool is absent or
unavailable in the active mode, ask the same three choices in one concise message and wait for the
answer. Do not attempt a known-unavailable tool and do not silently infer paid subscriptions.
Inspect PARABLE_CLIPROXY_BIN and PATH for an existing CLIProxyAPI executable. If none exists,
ask one more native yes/no question for consent to download source, install any missing build
prerequisite, and build Parable's pinned patched proxy. Prefer a user-local prerequisite install;
keep any Parable-owned data root at mode 0700; request separate approval before any privileged
system change. Stop cleanly if consent is denied.
Build the canonical vendor list with claude always present and optional chatgpt, xai, and
kimi from those answers.
What ships with it
25 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- parable.sh 4.9 KB runs code
- references/config.md 16 KB
- references/parable.example.toml 6.5 KB
- references/providers.md 21 KB
- references/review-prompt.md 1.8 KB
- references/routing.md 6.0 KB
- runtime/bin/parable.js 9.9 KB runs code
- runtime/lib/onboarding.js 81 KB runs code
- runtime/patches/cliproxyapi-v7.2.131-claude-effort.patch 14 KB
- runtime/VERSION 7 B
- runtime/welcome-plugin/.claude-plugin/plugin.json 163 B
- runtime/welcome-plugin/hooks/hooks.json 1.5 KB
- runtime/welcome-plugin/scripts/context_recovery.py 7.9 KB runs code
- runtime/welcome-plugin/scripts/model_guard.py 3.3 KB runs code
- runtime/welcome-plugin/scripts/welcome.py 301 B runs code
- scripts/parable_usage.py 19 KB runs code
- scripts/parable-config.sh 74 B runs code
- scripts/parable-list.sh 72 B runs code
- scripts/parable-resume.sh 74 B runs code
- scripts/parable-review.sh 74 B runs code
- scripts/parable-run.sh 71 B runs code
- scripts/parable-status.sh 74 B runs code
- scripts/parable-usage.sh 73 B runs code
- scripts/parable-verify.sh 74 B runs code
- scripts/parable.py 109 KB runs code
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.
- 2mo ago First seen · 363 lines · 115 tokens per session scan A c84c2d8e6a2f
parable is a skill published in the GitHub repository Parcha-ai/parcha-skills (126 stars, last pushed 15d ago), licensed MIT. It adds 115 tokens to every session and 5,801 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
parable compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| parableParcha-ai/parcha-skills | Skill | 126 | 15d ago | MIT | A | 115 | Claude Code |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | today | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | today | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 314k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |