{"id":51166526,"url":"https://github.com/teixeirazeus/fablize-for-hermes","last_synced_at":"2026-06-26T19:30:56.178Z","repository":{"id":365466724,"uuid":"1271709279","full_name":"teixeirazeus/fablize-for-hermes","owner":"teixeirazeus","description":" This project adapts fablize's verified procedures — verification grounding, multi-story evidence gating, investigation protocol, and early-stop prevention.","archived":false,"fork":false,"pushed_at":"2026-06-17T12:02:17.000Z","size":8023,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-17T14:04:26.885Z","etag":null,"topics":["agent-skills","agentskills","ai","ai-agents","artificial-intelligence","autonomous-agents","hermes-agent","hermes-ecosystem","hermes-skill","llm","nous-research","task-completion","verification-framework"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/teixeirazeus.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-06-17T00:05:20.000Z","updated_at":"2026-06-17T12:02:29.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/teixeirazeus/fablize-for-hermes","commit_stats":null,"previous_names":["teixeirazeus/fablize-for-hermes"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/teixeirazeus/fablize-for-hermes","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teixeirazeus%2Ffablize-for-hermes","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teixeirazeus%2Ffablize-for-hermes/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teixeirazeus%2Ffablize-for-hermes/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teixeirazeus%2Ffablize-for-hermes/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/teixeirazeus","download_url":"https://codeload.github.com/teixeirazeus/fablize-for-hermes/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/teixeirazeus%2Ffablize-for-hermes/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34831249,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-26T02:00:06.560Z","response_time":106,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent-skills","agentskills","ai","ai-agents","artificial-intelligence","autonomous-agents","hermes-agent","hermes-ecosystem","hermes-skill","llm","nous-research","task-completion","verification-framework"],"created_at":"2026-06-26T19:30:55.360Z","updated_at":"2026-06-26T19:30:56.170Z","avatar_url":"https://github.com/teixeirazeus.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003ca href=\"README.md\"\u003e\u003cstrong\u003eEnglish\u003c/strong\u003e\u003c/a\u003e ·\n  \u003ca href=\"README.zh.md\"\u003e简体中文\u003c/a\u003e ·\n  \u003ca href=\"README.pt.md\"\u003ePortuguês (BR)\u003c/a\u003e\n\u003c/p\u003e\n\n# fablize-for-hermes — run like Fable, on Hermes Agent\n![Banner](banner.png)\n[![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](https://opensource.org/licenses/MIT)\n\n\n\nA port of [fablize](https://github.com/fivetaku/fablize) for the [Hermes Agent](https://hermes-agent.nousresearch.com/)\necosystem. This project adapts fablize's verified procedures — verification grounding,\nmulti-story evidence gating, investigation protocol, and early-stop prevention — from\nClaude Code's plugin system to Hermes Agent's skill system.\n\n**fablize** makes any agent **see a task through to the end — with evidence and\nverification — as procedure, not as luck.**\n\n## The origin\n\n[fablize](https://github.com/fivetaku/fablize) was born from a controlled comparison\nof Claude Opus and \"Fable 5\" (19 A/B runs + 26 real sessions, ~1,500 tool calls).\nThe finding: on closed, answer-bearing work (code, logic, builds) the two models\nwere effectively tied. The gap was in *depth* — following an implication one step\nfurther — and that turned out to be **model capability**, not injectable.\n\nBut the **procedure** of good work — actually running what you build, decomposing\nmulti-step tasks with evidence checkpoints, investigating bugs systematically —\n*does transfer*. fablize ships only the procedures whose effect was verified.\n\n## What transfers and what doesn't\n\n| Trait | Transferable? | How fablize-for-hermes does it |\n|---|---|---|\n| Verification grounding (run \u0026 observe the artifact) | ✅ | Skill section 3 — run in real renderer, observe output, fix, re-run |\n| Multi-story completion + evidence gate | ✅ | `scripts/goals.py` — decompose, checkpoint, refuse completion without proof |\n| Systematic investigation (reproduce → hypotheses → causal chain) | ✅ | Skill section 2 + `packs/investigation-protocol.txt` |\n| Early-stop prevention | ✅ | Skill foundation rule: no promises without tool calls |\n| Out-of-spec defect discovery | ❌ not possible | Capability — the model finds it, or it doesn't |\n| Open-ended creative detail | ❌ not possible | Capability — shows only where there is no fixed answer |\n| Self-driven propagation depth | ❌ not possible | Capability — directed propagation transfers; self-started depth does not |\n\n## What's included\n\n### 1. Hermes Agent skill (`skills/fablize/SKILL.md`)\n\nThe core behavioral ruleset, structured as a Hermes Agent skill (YAML frontmatter +\nmarkdown). Covers:\n\n- **Foundation rules** — always-on: lead with outcome, ground claims in tool results,\n  no early stops, confirm destructive actions\n- **Multi-story loop** — decompose 2+ story tasks → complete with evidence →\n  verification gate on the final story\n- **Deep investigation** — reproduce → 3+ competing hypotheses → trace causal chain →\n  verify before/after → report rejected hypotheses\n- **Verification grounding** — run render/executable artifacts, observe, fix, re-run\n  before declaring done\n- **Capability ceiling** — escalate honestly when the model's ceiling is the blocker\n\n### 2. Goal engine (`scripts/goals.py`)\n\nA self-contained, stdlib-only Python script for multi-story decomposition with a\nverification gate. State persists in `~/.hermes/fablize/` and survives session death.\n\n```bash\ngoals.py create --brief \"Build a dashboard\" \\\n  --goal \"design::Create the HTML layout\" \\\n  --goal \"implement::Add chart visualizations\" \\\n  --goal \"verify::Run and verify the dashboard renders correctly\"\ngoals.py next\n# ... work this story ...\ngoals.py checkpoint --id G001 --status complete --evidence \"Layout created with sidebar and main panel\"\ngoals.py status\n```\n\n### 3. Packs — contextual discipline files\n\n- **`packs/verification-grounding-pack.txt`** — The grounding loop: run, observe,\n  fix, re-run. For HTML, SVG, games, UI, charts, animations.\n- **`packs/investigation-protocol.txt`** — The systematic debugging protocol:\n  reproduce, competing hypotheses, causal chain, verify before/after.\n\n### 4. Setup scripts (`setup/`)\n\n- `setup.sh [--skill-only|--local|--global]` — Install the skill into Hermes and\n  optionally inject the operating block into `AGENTS.md` or `CLAUDE.md`\n- `uninstall.sh [--skill-only|--local|--global|--all]` — Reverse the setup\n\n## Differences from original fablize\n\n| Aspect | fablize (original) | fablize-for-hermes |\n|---|---|---|\n| Platform | Claude Code plugin | Hermes Agent skill |\n| Routing | `UserPromptSubmit` hook via `hooks.json` | Skill content — agent loads and applies matching discipline |\n| Early-stop | `finish-the-work.sh` Stop hook (JSON transcript) | Foundation rule in skill — agent self-enforces |\n| State dir | `./.fablize/` (per-project) | `~/.hermes/fablize/` (Hermes-wide, survives sessions) |\n| Scripts | In `scripts/` — run from repo root | Same scripts, adapted for Hermes paths |\n| Setup | Injects into `CLAUDE.md` | Installs as Hermes skill + optionally injects block |\n\n## Install\n\n### Quick install (skill only)\n\n```bash\ngit clone https://github.com/teixeirazeus/fablize-for-hermes.git\ncd fablize-for-hermes\nbash setup/setup.sh --skill-only\n```\n\nThen load the skill in your Hermes session:\n\n```\nskill_view(name='fablize')\n```\n\n### Full install (skill + always-on block)\n\n```bash\nbash setup/setup.sh --local    # injects block into ./AGENTS.md\nbash setup/setup.sh --global   # injects block into ~/.claude/CLAUDE.md\n```\n\n### Uninstall\n\n```bash\nbash setup/uninstall.sh --all\n```\n\n## How it behaves\n\n- **Trigger:** load the `fablize` skill, or trigger inline when the agent detects\n  2+ stories, debugging, or render artifacts\n- **2+ stories** → decompose + verification gate via `goals.py`\n- **Debugging** → investigation protocol (reproduce → hypotheses → causal chain)\n- **Render artifact** → verification grounding (run → observe → fix → re-run)\n- **Hard task** → adaptive thinking + escalate to stronger model if stuck\n- **Capability ceiling** → escalate honestly (the harness does not fake capability)\n\n## Honest limits\n\n- It cannot raise model capability. Open-ended creative quality and self-driven\n  discovery are out of reach — that is a model-choice decision, not a harness one.\n- The effect numbers come from the original fablize small-family self-measurement.\n  The direction is solid; the decimals are not asserted.\n- The early-stop rule relies on agent self-enforcement, not a deterministic hook.\n  It is a guideline, not a guard.\n\n## License\n\nMIT — same as the original fablize. The upstream fablize by fivetaku is MIT licensed.\nThis port adapts the design and procedures to the Hermes Agent ecosystem.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fteixeirazeus%2Ffablize-for-hermes","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fteixeirazeus%2Ffablize-for-hermes","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fteixeirazeus%2Ffablize-for-hermes/lists"}