pluginbench 0.5.0__tar.gz

pluginbench-0.5.0/.gitignore

@@ -0,0 +1,17 @@
+.venv/
+node_modules/
+__pycache__/
+*.pyc
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.coverage
+htmlcov/
+dist/
+build/
+*.egg-info/
+runs/
+logs/
+.DS_Store
+.env
+auth.json

pluginbench-0.5.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Filip Gutica
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

pluginbench-0.5.0/PKG-INFO

@@ -0,0 +1,294 @@
+Metadata-Version: 2.5
+Name: pluginbench
+Version: 0.5.0
+Summary: Test whether skills help coding agents make fewer mistakes.
+Project-URL: Homepage, https://github.com/filipgutica/pluginbench
+Project-URL: Documentation, https://github.com/filipgutica/pluginbench/blob/main/docs/running-evaluations.md
+Project-URL: Repository, https://github.com/filipgutica/pluginbench
+Project-URL: Issues, https://github.com/filipgutica/pluginbench/issues
+Author: Filip Gutica
+License-Expression: Apache-2.0 AND BSD-3-Clause AND 0BSD
+License-File: LICENSE
+License-File: src/pluginbench/vendor/ECHARTS-LICENSE
+License-File: src/pluginbench/vendor/ECHARTS-NOTICE
+License-File: src/pluginbench/vendor/ZRENDER-LICENSE
+Keywords: benchmark,coding agents,evaluation,plugins,skills
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: MacOS
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Software Development :: Testing
+Requires-Python: >=3.12
+Requires-Dist: harbor==0.24.0
+Requires-Dist: pydantic<3,>=2.11
+Requires-Dist: pyyaml<7,>=6.0.2
+Requires-Dist: typer<1,>=0.16
+Provides-Extra: dev
+Requires-Dist: build<2,>=1.3; extra == 'dev'
+Requires-Dist: mypy<2,>=1.17; extra == 'dev'
+Requires-Dist: pytest<9,>=8.4; extra == 'dev'
+Requires-Dist: ruff<0.16,>=0.15; extra == 'dev'
+Requires-Dist: types-pyyaml<7,>=6.0.12; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# PluginBench
+
+**Are these skills, plugins, or instructions worth using?**
+
+PluginBench gives a coding agent the same tasks in several arms: one with no added guidance ("vanilla"), and one for each set of guidance you want to test.
+The harness, model, reasoning effort, and tasks stay fixed, so the guidance is the only difference.
+It then compares correctness, tokens, and the code each attempt added.
+
+## What you get
+
+- **Arms.** Vanilla is always included. Each *treatment* is a named set of instruction files, skills, or plugins that you declare in the config. Every treatment is compared with vanilla.
+- **Attempts.** Each task runs once per arm and repeat, in a fresh Docker container, through Harbor 0.24.0. The task's own verifier decides pass or fail.
+- **Measures.** You select them in the config:
+
+| Measurement | What it counts | Extra requirement |
+| --- | --- | --- |
+| `tokens` | Input, cached, output, and reasoning tokens, and coding time | None |
+| `lines` | Source and test lines added and removed | None |
+| `cyclomatic`, `cognitive` | Complexity the attempt added | An analyzer for the task's language |
+| `dead_code` | Unused code the attempt introduced | An analyzer for the task's language |
+| `coverage` | How much of the added code the attempt's own tests run | A coverage command for the task image |
+| `review` | Findings from an independent AI review with fixed criteria | A `reviewer` block in the config |
+
+- **Report.** `runs/<run>/report/report.html` compares the arms, with each task one click away. `results.json` holds the metrics and deltas against vanilla. `results.csv` holds each measurement. Failed attempts and missing measurements stay visible.
+
+When a task ships a reference solution (`solution/solution.patch`), code measures are also shown as a ratio to that reference. [What each measure means](docs/measurement-spec.md).
+
+## Requirements
+
+PluginBench runs on Linux or macOS. Every run needs:
+
+| Tool | Version | Why |
+| --- | --- | --- |
+| Python | 3.12 or later | Runs PluginBench |
+| Git | Any recent | Task sources and patches |
+| Docker | Linux containers on a local Unix socket, with Compose and Buildx | Each attempt and its verifier run in containers |
+| Codex CLI | Exactly 0.162.1 or 0.167.0, logged in | Coding agent. PluginBench checks the `codex` on your `PATH` and copies its login from `~/.codex/auth.json`. Other versions fail `check`. |
+
+Some measures and commands need more:
+
+| Needed for | Tool |
+| --- | --- |
+| JS/TS complexity and dead code | Node (this repository pins 22.20), Corepack, and Fallow 3.18.0 |
+| Python, Go, Rust analysis | Pinned analyzers; see [Measurements](docs/running-evaluations.md#measurements) |
+| `coverage` | A coverage command that the task image supports |
+| Analysis measures that run on the host, and `review` | Bubblewrap (`bwrap` on `PATH`) on Linux, or `sandbox-exec` on macOS |
+| `pluginbench tasks pick` | `fzf`, and for `--copy` a clipboard tool (`pbcopy`, `wl-copy`, `xclip`, or `xsel`) |
+
+## Install
+
+There is no registry release yet. Install from a clone:
+
+```sh
+git clone https://github.com/filipgutica/pluginbench.git
+cd pluginbench
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -e .
+pluginbench --version
+```
+
+## Quickstart
+
+This run uses one public DeepSWE task, vanilla only, and one repeat. It needs no private input.
+
+1. Save this as `evaluation.yaml`:
+
+   ```yaml
+   name: First run
+   runner:
+     harness: codex
+     model: gpt-6.1-sol
+     reasoning: medium
+   treatments: {}        # vanilla only
+   tasks:
+     urls:
+       - https://github.com/datacurve-ai/deep-swe/tree/0b9fabbb63b9104d678fe965e1632f2dd9eaa2ea/tasks/testem-per-launcher-reports
+   measurements: [tokens, lines]
+   repeats: 1
+   ```
+
+2. Run every check that needs no model:
+
+   ```sh
+   pluginbench check evaluation.yaml
+   ```
+
+   This downloads and prepares the task, checks Docker capacity, the Codex version, and the verifier, then prints a readiness table.
+   It shows whether each vanilla attempt is already cached, writes `runs/check-<timestamp>/check.json`, and exits with code 2 if the run would fail or lose data.
+   If it reports that `jobs` exceeds Docker capacity, add `jobs: 1` to the config or give Docker more CPU and memory.
+
+3. See the plan and the call count:
+
+   ```sh
+   pluginbench run evaluation.yaml --dry-run
+   ```
+
+   The `Coding attempts` line shows one attempt for this config. It counts attempts, not model requests: subagents are on by default, so one attempt can make several requests.
+
+4. Start the run. **This makes paid model calls with your Codex login.**
+
+   ```sh
+   pluginbench run evaluation.yaml
+   ```
+
+   `run` prints the plan and asks `Start evaluation?` (`--yes` skips the question).
+   It then runs the same checks as `check`, starts the run in the background, and prints the run directory, for example `runs/20261110T101500123456Z`.
+
+5. Follow it:
+
+   ```sh
+   pluginbench status runs/<run> --watch
+   ```
+
+   `status` shows the phase, attempts and outcomes per arm, tokens, vanilla reuse, and measures. It never writes.
+
+6. Open `runs/<run>/report/report.html` when the phase is `complete`.
+
+Paths under `runs/` are relative to the directory where you run the command. For a fuller template, `pluginbench init template.yaml` writes a starter config with `/path/to` placeholders to replace.
+
+## Test your own guidance
+
+A *treatment* is the guidance you want to test. It can be instruction files, skills, plugins, or a mix. Every treatment runs on the same tasks as vanilla, so you can see what each one changes.
+
+1. Put the guidance next to your config. Paths in the config are relative to the config file.
+
+   ```text
+   my-eval/
+   ├── evaluation.yaml
+   ├── AGENTS.md                     # instructions: UTF-8 .md or .txt files, 32 KiB in all
+   ├── skills/
+   │   └── write-tests/
+   │       └── SKILL.md              # a skill
+   └── review-plugin/
+       ├── .codex-plugin/
+       │   └── plugin.json           # a Codex plugin
+       └── skills/
+           └── self-review/
+               └── SKILL.md
+   ```
+
+   Each `SKILL.md` starts with YAML front matter that has a kebab-case `name` and a `description`:
+
+   ```markdown
+   ---
+   name: write-tests
+   description: Add focused tests for every behaviour you change.
+   ---
+
+   Write one test per changed behaviour. Run it and confirm it fails without your change.
+   ```
+
+   A plugin needs a manifest with a kebab-case `name`. Its skills go in `skills/`:
+
+   ```json
+   {
+     "name": "review-plugin",
+     "version": "0.1.0",
+     "description": "Review your own diff before you finish."
+   }
+   ```
+
+2. Declare one treatment per thing you want to compare:
+
+   ```yaml
+   name: Guidance comparison
+   runner:
+     harness: codex
+     model: gpt-6.1-sol
+     reasoning: medium
+   treatments:
+     instructions:
+       guidance: [./AGENTS.md]
+     test-skill:
+       skills: [./skills]            # one skill folder, or a folder of skills
+     review-plugin:
+       plugins: [./review-plugin]
+   tasks:
+     urls:
+       - https://github.com/datacurve-ai/deep-swe/tree/0b9fabbb63b9104d678fe965e1632f2dd9eaa2ea/tasks/testem-per-launcher-reports
+   measurements: [tokens, lines]
+   repeats: 1
+   ```
+
+   One treatment can also combine them, for example `guidance` and `skills` together.
+
+3. Check it, then look at the plan:
+
+   ```sh
+   pluginbench check evaluation.yaml
+   pluginbench run evaluation.yaml --dry-run
+   ```
+
+   The plan lists the arms `vanilla, instructions, test-skill, review-plugin` and 4 coding attempts: one per arm, task, and repeat. Add tasks and `repeats: 3` or more before you trust a difference.
+
+4. Run it with `pluginbench run evaluation.yaml`, as in the quickstart. The report compares each treatment with vanilla.
+
+Treatments need the Codex harness. Instruction files become the agent's private `AGENTS.md`; they are never added to the task prompt or to the repository. Skills and plugins are copied whole, including their scripts, except Git internals; symlinks, special files, and `.env` files are refused. Skill dependencies, MCP servers, apps, custom agents, commands, plugin dependencies, and hooks other than synchronous command hooks are not supported yet, and a treatment that declares one stops the run before any model call. Before coding, PluginBench checks that Codex found every declared skill. That proves the skill was available, not that the agent used it. [Native treatments](docs/running-evaluations.md#native-treatments).
+
+## Commands
+
+Every setting lives in the config. Flags only control how a command runs.
+
+| Command | What it does | Model calls |
+| --- | --- | --- |
+| `init [path]` | Write a starter config; never overwrites | No |
+| `check <config>` | Prepare tasks and run every check that needs no model | No |
+| `run <config>` | Start a run; `--dry-run` shows the plan, `--foreground` stays in the terminal, `--notify` sends desktop notifications | Yes |
+| `run <run-dir>` | Resume a stopped or crashed run | Yes |
+| `status <run-dir>` | Show progress; `--watch` redraws, `--json` prints one object | No |
+| `stop <run-dir>` | Stop the run's own process group | No |
+| `report <run-dir>` | Rebuild the report from saved evidence | No |
+| `review <run-dir> --revision <name>` | Review every submission again with the same reviewer | Yes |
+| `compare <run-dir>... --gates <file>` | Print measure and gate tables across runs | No |
+| `tasks pick --dataset <name>` | Choose tasks with fzf and print a pinned `tasks:` block | No |
+| `judge prepare`, `judge run`, `judge tally` | Blind pairwise judging of two arms | `judge run` only |
+
+## Configure an evaluation
+
+**Tasks.** Use one source: a Harbor `dataset` with `version` and `ids`, GitHub task-folder `urls`, or a local `source` with `ids`. PluginBench accepts single-step Harbor tasks with a Linux CPU environment and a separate verifier. To browse DeepSWE, run `pluginbench tasks pick --dataset datacurve/deep-swe-1-1 --copy`, press Tab to select several tasks, and paste the YAML it prints. [Choose inputs](docs/running-evaluations.md#choose-inputs).
+
+**Treatments.** Each treatment lists `guidance` (instruction files), `skills`, or `plugins`; see [Test your own guidance](#test-your-own-guidance). The task prompt is the same for every arm.
+
+**Measurements.** Before any paid call, PluginBench runs each selected collector twice on the starting code. A measure with no collector for a task is `not_configured`, and one whose collector fails its probe is `broken`. That task is then not measured for it, and the run continues, as long as another task can measure it. A measure that is ready on no task stops the run. With `measurement_settings.require_ready: true`, any such task stops the run. Selecting `review` requires a `reviewer` block. [Measurements](docs/running-evaluations.md#measurements).
+
+**Concurrency.** `jobs` defaults to 2 and accepts 1 to 4. A value above Docker capacity fails the check; PluginBench does not lower it. [Concurrency and resume](docs/running-evaluations.md#concurrency-and-resume).
+
+## Stop, resume, and reuse
+
+- `pluginbench stop runs/<run>` stops new attempts. With `jobs` above 1, attempts already coding finish and are kept. With `jobs: 1`, the attempt in progress is interrupted, and a resume codes its slot again with a new attempt.
+- `pluginbench run runs/<run>` resumes. A run that was killed, for example by a host shutdown, shows `interrupted`, and `status` prints this command. Complete attempts are kept and never coded again. An attempt cut off mid-call is recorded as `discarded-interrupted` and a new attempt takes its slot. Containers the killed attempt left running are stopped before coding resumes, and `status` names them until then. Resume cannot repair a failing solution.
+- If Codex fails before its first model turn, for example during a network outage, the run stops instead of scoring the attempt. Resume when Codex is reachable.
+- Vanilla attempts are cached in `$XDG_CACHE_HOME/pluginbench` (default `~/.cache/pluginbench`) and reused by any later run with the same task, model, and settings. Grading and measures run again. Reports mark reused attempts. [Baselines and reports](docs/running-evaluations.md#baselines-and-reports).
+
+## Compare and judge
+
+- `pluginbench compare` reads saved reports and applies gates you define, such as "no fewer passes than vanilla" or "at most 1.3 times the tokens". [Compare runs](docs/running-evaluations.md#compare-runs).
+- `pluginbench judge` asks blind judges to choose between two arms' submissions for the same task and repeat. [Judging](docs/running-evaluations.md#judging).
+
+## Limits
+
+- A pass means the task's verifier passed. Task checks do not prove that every behavior is preserved.
+- Discovery proves that a skill was available, not that the model used it.
+- Smaller numbers alone do not prove better work. Coverage does not prove that assertions are useful.
+- Unknown values are never shown as zero.
+- Reused vanilla attempts are not fresh independent samples.
+- A measure that is `broken` on a task is not measured for that task. `check` warns, and the report shows it as unavailable there. With `measurement_settings.require_ready: true`, the check stops instead. A measure that is ready on no task always stops the run.
+- PluginBench executes model-generated code. Use trusted tasks, images, and treatments on a trusted machine. See [Security](SECURITY.md).
+
+## More
+
+- [Evaluation details](docs/running-evaluations.md): every setting, check, and output
+- [Config reference](docs/config-reference.md): every config key, its type, default, and meaning
+- [Metrics](docs/metrics.md): what each measure means and its name in a gates file
+- [Reproduce a result](docs/reproduce.md): what a run pins, and a worked example
+- [Measurement specification](docs/measurement-spec.md)
+- [Security](SECURITY.md)
+- [License: Apache-2.0](LICENSE)