experience-memory
Save successful action tips as soon as they work and retrieve them for later actions under matching observable conditions; distill reusable procedures into…
Skill by robot-agent-exploration
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/robot-agent-exploration-experience-memory
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/experience-memory
What's inside
- Skill name
experience-memory- Description
- Save successful action tips as soon as they work and retrieve them for later actions under matching observable conditions; distill reusable procedures into learned skills with evidence. Use while controlling the robot, when building automation, and at the start of work or after compaction.
- License
- Apache-2.0
- Instructions
- 54 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- format.md · references/
SKILL.md (first lines)
# Experience memory `memory/` belongs to this task and persists across episodes, runs and context compaction. Keep learned procedures under `memory/skills/`, apart from the provided skills. ## Retrieve At the start of work or after compaction, read `memory/INDEX.md` and `memory/TIPS.md` if present, and open only the learned skills whose conditions match the current scene. Open linked evidence only when applicability or outcome is uncertain; do not load the whole history. Current perception and the live tool schemas take precedence over anything remembered. ## Record tips immediately - As soon as an action produces a visibly successful, useful result, add or update one row in `memory/TIPS.md`. A visible local result is enough; task-level confirmation can be added later. - Record the task phase, the observable preconditions, the tool and arguments or the reusable pose relation, the visible effect, and the saved frame paths the tools returned. Say what made it work, so later attempts skip the exploration and the superseded failures. - Refine the matching row when the same maneuver improves instead of adding variants. - Scenes drift between attempts (objects moved back by hand, lighting, layouts): record the conditions under which a tip held, not only its numbers. ## Distill Promote a tip to `memory/skills/<slug>/SKILL.md` plus `evidence.json` when it shows a reusable motion, contact strategy, recovery or completed procedure. Keep the procedure short and relative to observed features; keep attempt-specific values and evidence links in `evidence.json`. Encode poses as the robot's pose-record conventions define them, when [robot-interface](../robot-interface/SKILL.md) has them. Update `memory/INDEX.md` with the skill path, tags, confirmation level and the last episode or date. Mark `task-confirmed` only when the harness's trusted check accepted an attempt the procedure contributed to, or every `TASK.md` condition was checked in fresh images; otherwise `observed`. Revise a skill when later evidence contradicts it, and keep its provenance. Read [references/format.md](references/format.md) when creating or updating a learned skill. When a procedure has worked unattended and a persistent library exists, also contribute it there ([continual-learning](../continual-learning/SKILL.md)). ## Code that runs without memory A controller run for evaluation may not see the editable workspace: put every procedure it needs into its code before freezing it. When you continue from carried notes on a new scene, evidence paths may no longer resolve: rely on the stated conditions and pose recipes, replay relative recipes from the current measured state, and reacquire every point a recipe anchors to.
From robot-agent-exploration/rax, Apache-2.0.
More from robot-agent-exploration/rax
- 0.793score
object-localization
robot-agent-exploration
Locate visible object points in the robot's frame from the current observation, choosing the method the observation profile allows (depth with camera…
SkillAFreshness: A★ 184Claude Code +9 - 0.793score
autonomous-control
robot-agent-exploration
Turn what worked by hand into a step-wise Python controller, controller.py:main(ctx), that re-perceives the scene, completes the task from a fresh start and…
SkillAFreshness: A★ 184Claude Code +9 - 0.790score
multi-agent-exploration
robot-agent-exploration
Divide exploration among subagents with explicit robot or environment ownership and complementary motion or subtask goals, and combine their results.
SkillAFreshness: A★ 184Claude Code +9 - 0.790score
continual-learning
robot-agent-exploration
Read, run and grow a persistent skill library of verified, executable sub-task skills that carries what agents learn across sessions and tasks on the same…
SkillAFreshness: A★ 184Claude Code +9 - 0.788score
camera-inspection
robot-agent-exploration
Get a closer or different view of small, occluded or out-of-view task details by moving a camera the robot carries (a wrist camera on an empty arm, a pan-tilt…
SkillAFreshness: A★ 184Claude Code +9 - 0.788score
multi-env-exploration
robot-agent-exploration
Select and operate parallel exploration environments, keep their independent scenes apart, and share one episode budget.
SkillAFreshness: A★ 184Claude Code +9 - 0.788score
vlm-perception
robot-agent-exploration
Use the harness's vision-language model tool for image questions (identify, point at, segment, check) from direct tool calls or inside Python controllers…
SkillAFreshness: A★ 184Claude Code +9 - 0.784score
transfer-handoff
robot-agent-exploration
Write handoff/README.md, describing how to plan the motion for each subtask and object manipulation of this task, for a later session (or a later copy of you)…
SkillAFreshness: A★ 184Claude Code +9 - 0.784score
live-stream
robot-agent-exploration
The continuous, read-only recording of the robot's cameras and state over the recent past.
SkillAFreshness: A★ 184Claude Code +9 - 0.784score
manual-robot-control
robot-agent-exploration
Solve the task by choosing robot actions directly from current observations.
SkillAFreshness: A★ 184Claude Code +9 - 0.784score
in-context-action-learning
robot-agent-exploration
Infer task goals and a useful strategy from visual demonstrations (an image sequence of a human or robot completing the task, or a terminal-state image), and…
SkillAFreshness: A★ 184Claude Code +9 - 0.652score
robot-interface
robot-agent-exploration
Placeholder for the robot's own interface skill, which every universal skill links to for tools, frames, limits and safety.
SkillAFreshness: A★ 184Claude Code +9
Signals
- Stars
- 184
- Forks
- 12
- Open issues
- 0
- Last commit (this folder)
- Oct 14, 2026 (13 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/robot-agent-exploration/rax
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from robot-agent-exploration
- 0.793score
object-localization
robot-agent-exploration
Locate visible object points in the robot's frame from the current observation, choosing the method the observation profile allows (depth with camera…
SkillAFreshness: A★ 184Claude Code +9 - 0.793score
autonomous-control
robot-agent-exploration
Turn what worked by hand into a step-wise Python controller, controller.py:main(ctx), that re-perceives the scene, completes the task from a fresh start and…
SkillAFreshness: A★ 184Claude Code +9 - 0.790score
multi-agent-exploration
robot-agent-exploration
Divide exploration among subagents with explicit robot or environment ownership and complementary motion or subtask goals, and combine their results.
SkillAFreshness: A★ 184Claude Code +9 - 0.790score
continual-learning
robot-agent-exploration
Read, run and grow a persistent skill library of verified, executable sub-task skills that carries what agents learn across sessions and tasks on the same…
SkillAFreshness: A★ 184Claude Code +9 - 0.788score
camera-inspection
robot-agent-exploration
Get a closer or different view of small, occluded or out-of-view task details by moving a camera the robot carries (a wrist camera on an empty arm, a pan-tilt…
SkillAFreshness: A★ 184Claude Code +9 - 0.788score
multi-env-exploration
robot-agent-exploration
Select and operate parallel exploration environments, keep their independent scenes apart, and share one episode budget.
SkillAFreshness: A★ 184Claude Code +9
Cite this page
experience-memory skill. Agent Mart. https://agentmart.dev/skills/robot-agent-exploration-experience-memory (last updated 2026-10-24)