🔧 Update OpenCode development tooling

Expose selected agent skills as slash commands and pass invocation
arguments to their entry points.

Update the devenv OpenCode CLI to v2.0.16 with checksums for both
supported architectures.

AI-assisted-by: space-bunny-free

.agents/skills/create-commit/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: create-commit
 description: Stage, review, and commit files following Penpot commit conventions.
+slash: true
 ---
 
 # Skill: create-commit

.agents/skills/create-issue/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: create-issue
 description: Create or update GitHub issues (from PR, from draft body, retitle existing). Routes to the canonical flow in `mem:workflow/creating-issues`.
+slash: true
 ---
 
 # Skill: create-issue

.agents/skills/implement-plan/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: implement-plan
-description: Implementation flow — execute a ready plan from the session context: read the plan, detect the flow, then present the full picture (issue and branch to create or the branch to continue on, execution style, task checklist) and wait for confirmation. Default is every task with one final commit; on request ("step by step"), one task and one commit at a time with a pause after each; on request ("direct"), no issue and no branch — the commit lands on the current branch. Use it when the user asks to implement or execute a plan, in any phrasing.
+description: Implementation flow — execute a ready plan from the session context
+slash: true
 ---
 
 # Implement Plan
@@ -152,11 +153,6 @@ instruction from me overrides them):
   `/make-a-plan` by itself if the findings need one.
 - `/create-pr` — when the task is done and the branch is ready to merge.
 
-## User context
+## User input, overrides and additional context
 
-Extra context in the user's invocation (the message that triggered this
-skill) plays the role command arguments play elsewhere: `standalone`,
-`continue`, `direct` (`no branch` / `direct commit`), `no issue` /
-`without issue`, an explicit base such as `from origin/develop`, or
-`step by step` / `one commit per task` for the step-by-step execution
-mode. Modes combine freely, for example "standalone step by step".
+$ARGUMENTS

.agents/skills/make-a-plan/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: make-a-plan
-description: Planning flow — research the subject of this session, produce an implementation plan with the planner skill, resolve open questions with the user in plain language, and save the final plan to .agents/plans/. Use it when the user asks to plan, design, or break down a task, in any phrasing.
+description: Planning flow — research the subject of this session, produce an implementation plan with the planner skill, resolve open questions with the user in plain language, and save the final plan to .agents/plans/.
+slash: true
 ---
 
 # Make a Plan
@@ -106,9 +107,6 @@ the final response by suggesting the next steps, in this order:
 These are suggestions, not a required pipeline — any instruction from me
 overrides them (for example, asking you to implement the plan directly).
 
-## User context
+## User input, overrides and additional context
 
-Extra context in the user's invocation (the message that triggered this skill)
-plays the role command arguments play elsewhere: for example, `delegated` to
-hand the research and drafting to the `general` subagent, or corrections and
-feedback about a previous plan.
+$ARGUMENTS

.agents/skills/review-code/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: review-code
 description: Code review flow — review a diff, PR, or code change, delegating the review to a subagent that follows the code-review-criteria skill. Use it when the user asks to review code or a PR, in any phrasing.
+slash: true
 ---
 
 # Review Code
@@ -66,8 +67,6 @@ agent again.
 6. Skip generated files, lockfile-only changes, and unrelated modifications
    unless they introduce security risks.
 
-## User context
+## User input, overrides and additional context
 
-Extra context in the user's invocation (the message that triggered this skill)
-plays the role command arguments play elsewhere: for example, a PR number or
-URL, a commit range, specific files, or a different agent to run the review.
+$ARGUMENTS

.agents/skills/review-plan/SKILL.md

@@ -1,6 +1,7 @@
 ---
 name: review-plan
 description: Plan review flow — evaluate an implementation plan before it is executed, delegating the review to a subagent that follows the plan-review-criteria skill. Use it when the user asks to review a plan, in any phrasing.
+slash: true
 ---
 
 # Review Plan
@@ -64,8 +65,6 @@ agent again.
 5. Judge the plan as the implementer would: every task executable without
    guessing, ordering follows the dependency graph, risks named.
 
-## User context
+## User input, overrides and additional context
 
-Extra context in the user's invocation (the message that triggered this skill)
-plays the role command arguments play elsewhere: for example, a plan file path
-to review, or a different agent to run the review.
+$ARGUMENTS

docker/devenv/Dockerfile

@@ -136,18 +136,18 @@ RUN set -ex; \
 
 FROM base AS setup-opencode2
 
-ENV OPENCODE2_VERSION=2.0.15
+ENV OPENCODE2_VERSION=2.0.16
 
 RUN set -eux; \
     ARCH="$(dpkg --print-architecture)"; \
     case "${ARCH}" in \
        aarch64|arm64) \
          BINARY_URL="https://registry.npmjs.org/@opencode/cli-linux-arm64/-/cli-linux-arm64-${OPENCODE2_VERSION}.tgz"; \
-         ESUM='add6b6f92c05db3fb31b7ccf30f42cab38e72475e707f78c3d3c5e52ce60e22f'; \
+         ESUM='a132a041968b0fcea18824f2f1996f1494c26ad6e4c5be5ca5b9b96a2bb7922e'; \
          ;; \
        amd64|x86_64) \
          BINARY_URL="https://registry.npmjs.org/@opencode/cli-linux-x64/-/cli-linux-x64-${OPENCODE2_VERSION}.tgz"; \
-         ESUM='5ba4e87d9ee5dd5ce3aa45acdba1c6d37acdd282d450d6b65fb45226d6da44cc'; \
+         ESUM='08221eb6a781354e9be3b3918ab5b7d91aff2f20739e7e2e0af81bcdc0822c8a'; \
          ;; \
        *) \
          echo "Unsupported arch: ${ARCH}"; exit 1; \