Executing Plans
Overview
Load plan, review critically, execute all tasks, report when complete.
Announce at start: "I'm using the executing-plans skill to implement this plan."
For non-trivial plan execution, include Aegis Visibility in natural prose:
name why Aegis is keeping the current slice tied to the approved plan,
checkpoint, drift check, pre-edit governance, or verification boundary. This
visibility belongs to the active execution workflow; do not replace it with a
generic used-skills log.
When execution reaches completion, do not invent a separate final report shape
for this workflow. Pass plan adherence, checkpoint/drift status, verification,
complexity, and residual risk into verification-before-completion so the
user-facing closeout uses the unified Aegis impact/safety receipt.
Note: Tell your human partner that Aegis works much better with access to subagents. The quality of its work will be significantly higher if run on a platform with subagent support (such as Claude Code or Codex). If subagents are available, use aegis:subagent-driven-development instead of this skill.
The Process
Step 1: Load and Review Plan
- Read plan file
- If the plan or active checkpoint includes an
Execution Readiness View,
read it before implementation and compare the plan against its intent lock,
scope fence, baseline lock, owner / contract constraints, compatibility
boundary, retirement boundary, test obligations, review gates, drift /
rewind rules, and evidence required before completion.
- Review critically - identify any questions or concerns about the plan
- If the view contradicts the plan, baseline, or current worktree evidence,
return to plan review or refresh the advisory handoff before editing.
- Run the TDD Route Guard before implementation:
- confirm the plan records
Mode, Decision, Strict authority, Test posture, and verification;
- treat plan approval, a bug label, architecture risk, contract risk, and
shared-module wording as non-authoritative for strict TDD;
- allow
Write failing test, Verify RED, GREEN, or REFACTOR task
steps only when Decision: strict and its strict authority are both
recorded;
- in
off, a missing record may be repaired only as Mode: off / Decision: skipped unless an explicit user/project strict request is present; this
record does not load test-driven-development;
- in
auto, a missing decision or a strict-looking task without strict
authority returns to plan review. Do not infer strict during execution.
- If concerns: Raise them with your human partner before starting
- If no concerns: Create TodoWrite and proceed
Step 1.5: Long-Task Checkpoint Setup
If the plan has multiple tasks, may span sessions, or includes architecture / contract / workflow changes:
- Announce: "I'm using the long-task-continuation skill to keep this plan checkpointed and drift-aware."
- Load aegis:long-task-continuation.
- Create the initial checkpoint from the plan:
- current todo
- active task
- completed tasks
- evidence refs
- blockers
- next step
- Before each task, restate the current checkpoint.
- After each task, update checkpoint, evidence refs, and drift check.
Before a verification-driven unplanned edit, read retained PatchShape,
CanonicalOwner, UpwardDrillSignal, outcome, and evidence refs. Route their
comparison with the candidate to systematic-debugging before editing; it
decides whether the directions converge. A proven independent canonical-owner
root stays on the normal plan path.
Step 2: Execute Tasks
For each task:
Mark as in_progress
Follow each step exactly (plan has bite-sized steps)
Before any new source-code path is added by a task, restate the plan's
Change Necessity or create a compact one if the plan failed to carry it
forward. Plan approval is not by itself proof that a new helper, small guard,
new branch, fallback, adapter, or owner is necessary.
Change Necessity:
- User-visible need:
- No-change / non-code option:
- Why code change is necessary:
- Minimum change boundary:
- Decision: no-change | docs/config-only | code-change | needs-clarification
If the decision is not code-change, pause execution and return to plan
review instead of editing. If the decision is code-change, carry the
minimum boundary into the edit and verification scope.
Before any non-trivial source edit, run the plan's
Pre-Edit Complexity Check or create a compact one:
Use using-aegis/references/complexity-governance.md for shared artifact
classes, pressure signals, and over-budget handling.
Complexity Budget:
- Artifact class:
- Target files / artifacts:
- Current pressure:
- Projected post-change pressure:
- Budget result: within-budget | at-risk | over-budget
- Planned governance:
Pre-Edit Complexity Check:
- Target edit file:
- Existing pressure signal:
- Safer edit boundary:
- Decision: edit-in-place | extract helper | add owner file | split task | pause for plan update
Pre-Edit Owner-Fit Decision:
- Edit intent: wiring-only | move-out / extract-first | local-fix-without-new-responsibility | new-responsibility | emergency / compatibility patch
- Owner fit:
- Safer edit boundary:
- Decision: edit-in-place | extract helper | add owner file | split task | pause for plan update
If the check contradicts the plan's file boundary, pause and return to plan
review instead of silently stuffing logic into an overloaded owner. If the
budget result is over-budget and the task does not also govern that
overrun, stop execution and return to plan review rather than pushing the
task through as if it were still atomic.
When the target edit file is over-budget or mixed-purpose,
new-responsibility must not be added in place by default. wiring-only,
move-out / extract-first, and local-fix-without-new-responsibility may
proceed only when they do not add a new responsibility and the verification
boundary is clear. emergency / compatibility patch requires residual risk
and a retirement trigger.
Run verifications as specified
Update TodoCheckpointDraft and DriftCheckDraft before marking the task completed.
When an Execution Readiness View exists, the drift check must explicitly
compare the active slice against the view's intent lock, scope fence,
baseline lock, compatibility boundary, retirement boundary, test
obligations, and review gates.
Mark as completed
Step 3: Complete Development
After all tasks complete and verified:
- Announce: "I'm using the finishing-a-development-branch skill to complete this work."
- REQUIRED SUB-SKILL: Use aegis:finishing-a-development-branch
- Follow that skill to verify tests, present options, execute choice
When to Stop and Ask for Help
STOP executing immediately when:
- Hit a blocker (missing dependency, test fails, instruction unclear)
- Plan has critical gaps preventing starting
- You don't understand an instruction
- Verification fails repeatedly
Ask for clarification rather than guessing.
When to Revisit Earlier Steps
Return to Review (Step 1) when:
- Partner updates the plan based on your feedback
- Fundamental approach needs rethinking
Don't force through blockers - stop and ask.
Remember
- Review plan critically first
- Follow plan steps exactly
- Don't skip verifications
- Reference skills when plan says to
- Stop when blocked, don't guess
- Never start implementation on main/master branch without explicit user consent
Integration
Required workflow skills:
- aegis:using-git-worktrees - REQUIRED: Set up isolated workspace before starting
- aegis:writing-plans - Creates the plan this skill executes
- aegis:finishing-a-development-branch - Complete development after all tasks
Source: hashgraph-online/awesome-codex-plugins → plugins/GanyuanRan/Aegis/skills/executing-plans/SKILL.md
1---
2name: executing-plans-3
3description: "Use when you have a written implementation plan to execute in a separate session with review checkpoints"
4category: general-purpose
5source_repo: hashgraph-online/awesome-codex-plugins
6source_path: "plugins/GanyuanRan/Aegis/skills/executing-plans/SKILL.md"
7source_url: https://github.com/hashgraph-online/awesome-codex-plugins/blob/HEAD/plugins/GanyuanRan/Aegis/skills/executing-plans/SKILL.md
8metadata:
9 source: https://skillmd.com/skills/hashgraph-online/executing-plans-3
10---
11
12
13# Executing Plans
14
15## Overview
16
17Load plan, review critically, execute all tasks, report when complete.
18
19**Announce at start:** "I'm using the executing-plans skill to implement this plan."
20
21For non-trivial plan execution, include `Aegis Visibility` in natural prose:
22name why Aegis is keeping the current slice tied to the approved plan,
23checkpoint, drift check, pre-edit governance, or verification boundary. This
24visibility belongs to the active execution workflow; do not replace it with a
25generic used-skills log.
26
27When execution reaches completion, do not invent a separate final report shape
28for this workflow. Pass plan adherence, checkpoint/drift status, verification,
29complexity, and residual risk into `verification-before-completion` so the
30user-facing closeout uses the unified Aegis impact/safety receipt.
31
32**Note:** Tell your human partner that Aegis works much better with access to subagents. The quality of its work will be significantly higher if run on a platform with subagent support (such as Claude Code or Codex). If subagents are available, use aegis:subagent-driven-development instead of this skill.
33
34## The Process
35
36### Step 1: Load and Review Plan
371. Read plan file
382. If the plan or active checkpoint includes an `Execution Readiness View`,
39 read it before implementation and compare the plan against its intent lock,
40 scope fence, baseline lock, owner / contract constraints, compatibility
41 boundary, retirement boundary, test obligations, review gates, drift /
42 rewind rules, and evidence required before completion.
433. Review critically - identify any questions or concerns about the plan
444. If the view contradicts the plan, baseline, or current worktree evidence,
45 return to plan review or refresh the advisory handoff before editing.
465. Run the TDD Route Guard before implementation:
47 - confirm the plan records `Mode`, `Decision`, `Strict authority`, `Test
48 posture`, and verification;
49 - treat plan approval, a bug label, architecture risk, contract risk, and
50 shared-module wording as non-authoritative for strict TDD;
51 - allow `Write failing test`, `Verify RED`, `GREEN`, or `REFACTOR` task
52 steps only when `Decision: strict` and its strict authority are both
53 recorded;
54 - in `off`, a missing record may be repaired only as `Mode: off / Decision:
55 skipped` unless an explicit user/project strict request is present; this
56 record does not load `test-driven-development`;
57 - in `auto`, a missing decision or a strict-looking task without strict
58 authority returns to plan review. Do not infer `strict` during execution.
596. If concerns: Raise them with your human partner before starting
607. If no concerns: Create TodoWrite and proceed
61
62### Step 1.5: Long-Task Checkpoint Setup
63
64If the plan has multiple tasks, may span sessions, or includes architecture / contract / workflow changes:
65
661. Announce: "I'm using the long-task-continuation skill to keep this plan checkpointed and drift-aware."
672. Load aegis:long-task-continuation.
683. Create the initial checkpoint from the plan:
69 - current todo
70 - active task
71 - completed tasks
72 - evidence refs
73 - blockers
74 - next step
754. Before each task, restate the current checkpoint.
765. After each task, update checkpoint, evidence refs, and drift check.
77
78Before a verification-driven unplanned edit, read retained `PatchShape`,
79`CanonicalOwner`, `UpwardDrillSignal`, outcome, and evidence refs. Route their
80comparison with the candidate to `systematic-debugging` before editing; it
81decides whether the directions converge. A proven independent canonical-owner
82root stays on the normal plan path.
83
84### Step 2: Execute Tasks
85
86For each task:
871. Mark as in_progress
882. Follow each step exactly (plan has bite-sized steps)
893. Before any new source-code path is added by a task, restate the plan's
90 `Change Necessity` or create a compact one if the plan failed to carry it
91 forward. Plan approval is not by itself proof that a new helper, small guard,
92 new branch, fallback, adapter, or owner is necessary.
93
94 ```text
95 Change Necessity:
96 - User-visible need:
97 - No-change / non-code option:
98 - Why code change is necessary:
99 - Minimum change boundary:
100 - Decision: no-change | docs/config-only | code-change | needs-clarification
101 ```
102
103 If the decision is not `code-change`, pause execution and return to plan
104 review instead of editing. If the decision is `code-change`, carry the
105 minimum boundary into the edit and verification scope.
1064. Before any non-trivial source edit, run the plan's
107 `Pre-Edit Complexity Check` or create a compact one:
108
109 Use `using-aegis/references/complexity-governance.md` for shared artifact
110 classes, pressure signals, and `over-budget` handling.
111
112 ```text
113 Complexity Budget:
114 - Artifact class:
115 - Target files / artifacts:
116 - Current pressure:
117 - Projected post-change pressure:
118 - Budget result: within-budget | at-risk | over-budget
119 - Planned governance:
120
121 Pre-Edit Complexity Check:
122 - Target edit file:
123 - Existing pressure signal:
124 - Safer edit boundary:
125 - Decision: edit-in-place | extract helper | add owner file | split task | pause for plan update
126
127 Pre-Edit Owner-Fit Decision:
128 - Edit intent: wiring-only | move-out / extract-first | local-fix-without-new-responsibility | new-responsibility | emergency / compatibility patch
129 - Owner fit:
130 - Safer edit boundary:
131 - Decision: edit-in-place | extract helper | add owner file | split task | pause for plan update
132 ```
133
134 If the check contradicts the plan's file boundary, pause and return to plan
135 review instead of silently stuffing logic into an overloaded owner. If the
136 budget result is `over-budget` and the task does not also govern that
137 overrun, stop execution and return to plan review rather than pushing the
138 task through as if it were still atomic.
139 When the target edit file is over-budget or mixed-purpose,
140 `new-responsibility` must not be added in place by default. `wiring-only`,
141 `move-out / extract-first`, and `local-fix-without-new-responsibility` may
142 proceed only when they do not add a new responsibility and the verification
143 boundary is clear. `emergency / compatibility patch` requires residual risk
144 and a retirement trigger.
1455. Run verifications as specified
1466. Update `TodoCheckpointDraft` and `DriftCheckDraft` before marking the task completed.
147 When an `Execution Readiness View` exists, the drift check must explicitly
148 compare the active slice against the view's intent lock, scope fence,
149 baseline lock, compatibility boundary, retirement boundary, test
150 obligations, and review gates.
1517. Mark as completed
152
153### Step 3: Complete Development
154
155After all tasks complete and verified:
156- Announce: "I'm using the finishing-a-development-branch skill to complete this work."
157- **REQUIRED SUB-SKILL:** Use aegis:finishing-a-development-branch
158- Follow that skill to verify tests, present options, execute choice
159
160## When to Stop and Ask for Help
161
162**STOP executing immediately when:**
163- Hit a blocker (missing dependency, test fails, instruction unclear)
164- Plan has critical gaps preventing starting
165- You don't understand an instruction
166- Verification fails repeatedly
167
168**Ask for clarification rather than guessing.**
169
170## When to Revisit Earlier Steps
171
172**Return to Review (Step 1) when:**
173- Partner updates the plan based on your feedback
174- Fundamental approach needs rethinking
175
176**Don't force through blockers** - stop and ask.
177
178## Remember
179- Review plan critically first
180- Follow plan steps exactly
181- Don't skip verifications
182- Reference skills when plan says to
183- Stop when blocked, don't guess
184- Never start implementation on main/master branch without explicit user consent
185
186## Integration
187
188**Required workflow skills:**
189- **aegis:using-git-worktrees** - REQUIRED: Set up isolated workspace before starting
190- **aegis:writing-plans** - Creates the plan this skill executes
191- **aegis:finishing-a-development-branch** - Complete development after all tasks
192
193---
194
195**Source:** [`hashgraph-online/awesome-codex-plugins`](https://github.com/hashgraph-online/awesome-codex-plugins) → `plugins/GanyuanRan/Aegis/skills/executing-plans/SKILL.md`