Files
worklane/docs/agent-assets.md
T
Ansho Rei 19c666b5dc Close a review of landed work by filing its remaining issues as tickets, and bump to 0.33.10
Work already on the default branch cannot go back to committed without
reopening shipped code. The reviewer files each remaining issue as a new
ticket, related issues together, comments why the ticket is closing, and
moves it to done.
2026-11-06 18:43:07 +01:00

33 lines
12 KiB
Markdown

# Shipping the Agent Assets
The documents and the opencode tool file the binary carries, and how they are installed.
- The skill is embedded with `include_str!` from `assets/SKILL.md`, not copied by an install script. `cargo install` places binaries and nothing else, so an installed document that is not in the binary is a document that drifts from it
- **`assets/SKILL.md` is the only worklane skill document in this repo, and nothing may reintroduce a second.** There was one: `.opencode/skill/worklane-kanban-skill/SKILL.md`, 223 lines of the same document that nothing shipped and no test checked, and it drifted exactly as far as a copy nobody reads does. Its frontmatter still called the skill `worklane_kanban`, its examples still used short flags the CLI had stopped accepting, and it knew nothing about `relate`, `campaign`, `chown`, `check` or the four-status flow. It was **not** inert while it existed: `opencode debug skill` listed it, so an agent working in this repo saw two worklane skills and could load either. Deleted by b2c5531. A copy under `.opencode/`, in a second assets file, or in the README is the same mistake; the document has one home and `install-skill` is how it travels
- **One install path serves both agents:** `~/.claude/skills/worklane/SKILL.md`. Claude Code reads it, and opencode loads that directory as an external skill source alongside its own `~/.config/opencode/skill/`. Verified against opencode 1.15.10, whose built-in `customize-opencode` skill documents the table and whose `opencode debug skill` lists what it found there. Do not add the second path: it is two copies of one document, free to drift, for no gain
- `Asset` carries the content, the path under its root and the word used in messages, so `install-tool` adds an entry and its own root rather than a second copy of the compare-and-refuse logic. That is why the module is `install.rs` and not `skill.rs`
- `install()` takes the root as an argument. That is what lets the tests install into a temp directory instead of the home of whoever runs the suite, and the integration test redirects `HOME` for the same reason
- **A modified copy is refused, not overwritten.** Someone may have edited it on purpose, and the message names `--print` so it can be diffed and `--force` so it can be replaced. A read error is not treated as absent either: the write reports the real reason
- Three tests guard the document rather than the mechanism, and they are the ones that earn their keep: the frontmatter has to parse (both agents silently ignore a document without it), every `worklane <x>` the skill shows has to be a real subcommand, and every subcommand outside `NOT_FOR_AGENTS` has to appear in the skill. The last one catches a new command that nothing tells an agent about
- `install-skill` looks for no board. The skill is what teaches an agent worklane, so needing a `.worklane` to install it would put it out of reach exactly when it is wanted
- The **agent files** under `.opencode/` are the user's, not this repo's to ship. Do not embed or install them. `reviewing-agent.md` there is the prior art the review skill grew out of and is deliberately **left in place**: its content is now in `assets/REVIEW.md`, but deleting a file under `.opencode/` is not this repo's call. Whether the stale copies there go is b2c5531's question. The tool file is the exception and no longer lives there: it was `.opencode/tools/worklane.ts`, which meant it existed only inside a checkout, and it is now `assets/TOOL.ts` and shipped like every other asset
- **`install-skill` writes every shipped document, and the name narrows it.** One document teaches the CLI and the others teach a job done with it, so a person who learned the invocation before a document existed still receives it. Requiring a name per document would mean a new skill silently never arrives for anyone who does not read the changelog
- **`--print` requires the name.** It exists so a modified copy can be diffed, and a diff is against one document; printing them concatenated would be neither of them. The unnamed case errors with the list of names rather than guessing, the same choice the `comment --` body makes
- The refusal message names the document it is about, `install-skill worklane --print`, since with several installed "compare with --print" does not say compare what
- **A refused document does not stop the ones behind it.** The loop reports an outcome per document and the "pick it up on their next session" line is printed once, and only when something was actually written
- **A role skill is its own document, not a section of `SKILL.md`.** The frontmatter description is the whole of what decides whether a skill loads, and one description cannot name both "anything touching a ticket" and "only when a review is asked for" without fitting neither. It is also why the review document does not restate the CLI: it says to read the `worklane` skill too
- The review skill's own rules are guarded by `the_review_skill_keeps_the_decisions_that_define_it`, because each of them is a line an edit could drop without breaking anything: **the reviewer never ticks a criterion** (a tick is the implementer's claim, unticking is the review disagreeing, and one mark for both loses the disagreement), **never moves the ticket** (`mv` belongs to whoever does the work, the reviewer's half is the recommendation) **unless the work is already on master** (then each finding gets its own ticket, related ones sharing one, and the reviewer moves the original to done once all of them are filed, after a comment saying which ticket holds each remaining issue, since a done ticket with unticked criteria otherwise reads as closed by mistake: sending landed work back to committed reopens shipped code, and leaving it in reviewing with nothing left to do only hides that the review is over), and **never reviews the working tree when no commit names the ticket** (it says what it searched for and stops, since an untraceable ticket in `reviewing` is itself the finding)
- **The plan skill is the one that can do damage, so its rules are the ones under test.** `set --tags` replaces the whole field, so a sweep that restates a ticket's tags incompletely deletes the rest, and a due date is neither validated nor clearable. `the_plan_skill_keeps_the_rules_that_bound_a_sweep` pins propose-then-apply and those two traps, because a document that loses them still reads like a plan
- `.opencode/agent/feature-planner-agent.md` is **deliberately not folded into `PLAN.md`**. It decomposes a feature request into new tickets, which is intake; `PLAN.md` orders the tickets that already exist. Merging them would give one document two jobs and a frontmatter description that fits neither, which is the same reason the review skill is its own document
- **`install-tool` is a second command, not a flag on `install-skill`, because the two artifacts are not alike.** A skill is inert markdown that any agent can read and it reaches both from one path; the tool is executable TypeScript that reaches opencode alone. Claude Code has no directory it scans for a tool definition at all: `claude mcp add` with a `stdio`, `sse` or `http` transport is the only way a tool enters its tool list, and its plugin system adopts `agents`, `hooks`, `monitors`, `workflows`, `output-styles` and `themes` and nothing named tools. So an MCP server, a different artifact, would be what reached it. Verified against Claude Code 2.1.238
- **`tools_root()` follows `XDG_CONFIG_HOME`, `skills_root()` does not.** `~/.claude` is not an XDG path and opencode's config root is: `opencode debug paths` reports `$XDG_CONFIG_HOME/opencode` with the variable set and `~/.config/opencode` without. An empty value counts as unset, the convention `colors_wanted` follows for `NO_COLOR`. `opencode_tool_dir` takes both variables as arguments for the reason `colors_wanted` does, since a threaded test binary cannot mutate the process environment to reach one branch at a time
- **Singular `tool`, and it was measured, not guessed.** opencode 1.15.10 loads `~/.config/opencode/tool/` and `~/.config/opencode/tools/` both: each was served in turn from a directory outside any project and `/experimental/tool/ids` registered every worklane tool the file carried, eleven at the time and twenty-one at the recheck after the wrapper caught up with the CLI. The singular is the spelling opencode's own documentation puts first, the way it writes `agent(s)` and `skill(s)`. `opencode serve` plus that endpoint is how to check this again; there is no `opencode debug tool` to ask
- **`worklane.ts` is the file name the tool ids are built from.** opencode names each tool as the file then the export, so the file's twenty-one `export const` lines become `worklane_add` through `worklane_tags_set`. Renaming the file renames every tool and every permission rule that mentions one, which is why `TOOL.relative` is pinned by a test rather than left as a detail
- **The tool needs nothing installed beside opencode**, which is worth stating because the file is executable and reads as though it wants a toolchain. It is loaded into opencode's own Bun runtime and the `@opencode-ai/plugin` import resolves there: a copy in a project with no `node_modules` anywhere above it registered all of its tools, on a machine with no `bun` on `PATH`. c092ea7's criterion 6 named those two as runtime requirements and it was wrong about both
- **The closing line differs from the skills', and that is not a wording choice.** opencode reads its config once at startup and never reloads it, so a running one keeps the tools it began with; the skills' "on their next session" would be a claim that is not true here
- **The wrapper walks clap in both directions now; until this ticket nothing did.** It was behind by four commands — `relate`, `campaign`, `assign` and `tags` had no tool, so an agent holding the tools could not reach campaigns or relations at all — and nothing compared the file to the CLI, which is how the gap opened and stayed open. `every_tool_the_wrapper_exports_is_a_real_subcommand` catches a tool for a command that has been renamed or dropped, which fails only when an agent finally calls it; `every_subcommand_outside_the_exceptions_has_a_tool` catches the other direction, a subcommand with no tool, through the same walk off clap the completion lists take. Its exception list carries a reason per entry rather than being bare, and one of those reasons was learned from the test's first run: a bare `worklane campaign` is refused by the CLI, its subcommand required, so there is no command for a `campaign` tool to wrap and the five `campaign_*` tools are the command
- **Installing it globally reproduces fb3dec6 for the wrapper.** Global tools apply in every project including this one, so once `install-tool` has run, an agent working on worklane uses the installed copy of `assets/TOOL.ts` and not the tree it is editing. Same trap as the `worklane` on `PATH`, second instance
- `every_agent_facing_subcommand_is_documented` stays on `SKILL` alone. It owns the CLI surface; a role document uses the handful of commands its job needs, and requiring the rest would fill a review document with `init` and campaigns to satisfy a test. The two checks that do run over every document are the frontmatter and that no document shows a command that does not exist
- **The README command table has the same guard, `every_subcommand_has_a_row_in_the_readme_command_table` in `command.rs`.** The skill and the table are the two places a command is discovered, one by an agent and one by a person, and the skill was the only one under test: `campaign` and `assign` shipped in 0.12.0 and were still absent from the table at 0.20.0, e6e3a1b, along with the `--campaign` flag on the `ls` row. It walks clap's own subcommands, one level of nesting included, against the `|` lines of `README.md`, so a new command fails the suite until the table lists it. `help` is excluded, being clap's. The tag subcommands are named in a `PENDING` list rather than skipped, since f7dd1a1 rewrites that row and emptying the list is part of closing it